The Lancet Infectious Diseases Commission

Antibiotic resistance—the need for global solutions

Ramanan Laxminarayan, Adriano Duse, Chand Wattal, Anita K M Zaidi, Heiman F L Wertheim, Nithima Sumpradit, Erika Vlieghe, Gabriel Levy Hara,
Ian M Gould, Herman Goossens, Christina Greko, Anthony D So, Maryam Bigdeli, Göran Tomson, Will Woodhouse, Eva Ombaka, Arturo Quizhpe Peralta,
Farah Naz Qamar, Fatima Mir, Sam Kariuki, Zulfiqar A Bhutta, Anthony Coates, Richard Bergstrom, Gerard D Wright, Eric D Brown, Otto Cars

The causes of antibiotic resistance are complex and include human behaviour at many levels of society; the consequences Lancet Infect Dis 2013;
affect everybody in the world. Similarities with climate change are evident. Many efforts have been made to describe the 13: 1057–98

many different facets of antibiotic resistance and the interventions needed to meet the challenge. However, coordinated Published Online
November 17, 2013
action is largely absent, especially at the political level, both nationally and internationally. Antibiotics paved the way for
http://dx.doi.org/10.1016/
unprecedented medical and societal developments, and are today indispensible in all health systems. Achievements in S1473-3099(13)70318-9
modern medicine, such as major surgery, organ transplantation, treatment of preterm babies, and cancer chemotherapy, This online publication
which we today take for granted, would not be possible without access to effective treatment for bacterial infections. has been corrected.
Within just a few years, we might be faced with dire setbacks, medically, socially, and economically, unless real and The corrected version first
appeared at thelancet.com/
unprecedented global coordinated actions are immediately taken. Here, we describe the global situation of antibiotic
infection on Nov 21, 2013
resistance, its major causes and consequences, and identify key areas in which action is urgently needed.
See Comment pages 1001,
1003, 1004, 1006, 1007, 1008,
Part 1: Global epidemiology of antibiotic Resistance has spread worldwide. Antibiotic-resistant and 1009
resistance and use gonorrhoea emerged in Vietnam in 1967,5 then spread to the Center for Disease Dynamics,
The rise of resistance Philippines, and finally the USA.6 NDM enzymes, first Economics and Policy,
The decreasing effectiveness of antibiotics in treating reported in 2008, are now found worldwide.7 The distri- Washington, DC, USA
(R Laxminarayan PhD);
common infections has quickened in recent years, and bution of resistance genes, such as Enterobacteriaceae- Princeton University, Princeton
with the arrival of untreatable strains of carbapenem- NJ, USA (R Laxminarayan);
resistant Enterobacteriaceae, we are at the dawn of a Public Health Foundation of
postantibiotic era.1 In high-income countries, continued Panel: Contributions to the Commission India, New Delhi, India
(R Laxminarayan); University of
high rates of antibiotic use in hospitals, the community, The Commission is a single document, each group of authors the Witwatersrand,
and agriculture have contributed to selection pressure that take responsibility for the text and views expressed in their Johannesburg, South Africa
has sustained resistant strains,2 forcing a shift to more individual parts. (A Duse MD); Sir Gangaram
Hospital, New Delhi, India
expensive and more broad-spectrum antibiotics. In low- Part 1: Global epidemiology of antibiotic resistance and use (C Wattal MD); Department of
income and middle-income countries (LMICs), antibiotic (page 1057) Paediatrics and Child Health,
use is increasing with rising incomes, high rates of Ramanan Laxminarayan, Adriano Duse, Chand Wattal, Division of Women and Child
hospitalisation, and high prevalence of hospital infections. Anita K M Zaidi Health, The Aga Khan
University Karachi, Sindh,
Resistance arises as a consequence of mutations in
Part 2: Getting out of the impasse (page 1061) Pakistan (A K M Zaidi MD,
microbes and selection pressure from antibiotic use that F N Qamar MBBS, F Mir PhD,
Heiman F L Wertheim, Nithima Sumpradit, Erika Vlieghe,
provides a competitive advantage for mutated strains. Prof Z A Bhutta PhD); Nuffield
Gabriel Levy Hara, Ian M Gould
Suboptimum antibiotic doses help stepwise selection of Department of Clinical
resistance. Resistance genes are borne on chromosomal, Part 3: Minimising the time to effective treatment—rapid Medicine, Centre for Tropical
diagnostic testing (page 1065) Diseases, University of Oxford,
and increasingly, on transmissible extrachromosomal Oxford, UK
elements. The resulting resistant clones—eg, meticillin- Herman Goossens (H F L Wertheim MD); Food and
resistant Staphylococcus aureus (MRSA) USA 300, Part 4: The interface between people and animals (page 1068) Drug Administration,
International Health Policy
Escherichia coli ST131, and Klebsiella ST258) are dis- Christina Greko
Program, Ministry of Public
seminated rapidly worldwide. This spread is facilitated by Part 5: The access and excess dilemma (page 1071) Health, Nonthaburi, Thailand
interspecies gene transmission, poor sanitation and Anthony D So, Maryam Bigdeli, Göran Tomson, Will Woodhouse, (N Sumpradit PhD); Institute of
hygiene in communities and hospitals, and the increasing Eva Ombaka, Arturo Quizhpe Peralta
Tropical Medicine, Antwerp,
Belgium and University
frequency of global, travel, trade, and disease transmission.
Part 6: Challenges of antibiotic resistance in weak health Hospital, Antwerp, Belgium
(E Vlieghe MD); Infectious
systems (page 1075)
Resistance is spreading worldwide Diseases Unit, Hospital
Farah Naz Qamar, Fatima Mir, Sam Kariuki, Zulfiqar A Bhutta Carlos G Durand, Buenos Aires,
Even before penicillin was introduced, resistant strains of
Part 7: Improving the interface between academics and the Argentina
bacteria had been detected.3 The selection pressure caused (Prof G Levy Hara MD);
by the use of millions of tonnes of antibiotics over the past pharmaceutical industry (page 1079)
Aberdeen Royal Infirmary,
75 years since antibiotics were introduced has made almost Anthony Coates, Richard Bergstrom Aberdeen, UK
all disease-causing bacteria resistant to antibiotics Part 8: Beyond antibiotics—alternative strategies for (I M Gould MBChB); Laboratory
of Medical Microbiology,
commonly used to treat them. The rapid evolution of prevention and treatment (page 1083) VAXINFECTIO, University of
bacterial resistance is clear in the case of β-lactamases class Gerard D Wright, Eric D Brown Antwerp, Antwerp, Belgium
of antibiotics. Nearly 1000 resistance-related β-lactamases Part 9: Call to action (page 1087) (H Goossens PhD); Department
that inactivate these antibiotics have been identified, a ten- of Animal Health and
Otto Cars, with contributions from all groups of authors Antimicrobial Strategies,
times increase since before 1990.4

www.thelancet.com/infection Vol 13 December 2013 1057

 The Lancet Infectious Diseases Commission

National Veterinary Institute, producing extended-spectrum β-lactamase (ESBL), NDM-1, presumed sepsis. Most worrying is the emergence of pan-
Uppsala, Sweden (C Greko PhD); and Klebsiella pneumoniae carbapenemase (KPC), indicates resistant untreatable carbapenem-resistant Entero-
Sanford School of Public Policy,
Duke University, Durham, NC,
the ease with which resistance can spread. Findings of a bacteriaceae and Acinetobacter spp infections associated
USA (Prof A D So MD, study8 done in New Delhi showed NDM-1-producing with high mortality in neonatal nurseries.16
W Woodhouse); Alliance for bacteria (including Shigella boydii and Vibrio cholera) in two In Pakistan, the emergence of pan-resistant bacterial
Health Policy and Systems (4%) of 50 drinking water samples and 51 (30%) of isolates such as Acinetobacter spp and carbapenem-
Research, WHO, Geneva,
Switzerland (M Bigdeli MPH);
171 seepage samples suggesting the possibility of acquiring resistant enterobacteria as causes of health-care associated
Departments of Learning, resistance outside health-care facilities. sepsis in hospitals is rendering these infections un-
Informatics, Management, Quinolone antibiotics in particular are an example of treatable.16–18 50–60% of community-acquired Gram-
Ethics and Public Health misadventure. These drugs are synthetic and so do not negative pathogens such as E coli associated with urinary
Sciences, Karolinska Institutet,
Stockholm, Sweden
arise in nature, yet 30 years after their widespread tract infections have become resistant to common oral
(Prof G Tomson MD); School of introduction resistance is epidemic.9 More specifically, antibiotics (eg, amoxicillin, cefixime, and ciprofloxacin),
Pharmacy, St John’s University whole genome studies suggest that quinolone resistance complicating outpatient management.
of Tanzania, Dodoma, was a crucial factor in the evolution of hospital MRSA.10 Between July 2010 and August 2011, 72% of 1294 viable
Tanzania(E Ombaka PhD);
Faculty of Medical Sciences,
Such examples of antibiotic-driven evolution go a long way K pneumoniae isolates from sentinel sites in South Africa
University of Cuenca, Cuenca, to explaining present epidemics of resistant health-care- had antibiograms suggestive of ESBL production.19 Com-
Ecuador associated infections.11 pounding this problem is the emergence of several
(Prof A Quizhpe Peralta MD);
In health-care settings, the spread of a resistant clone carbapenemase-resistance mechanisms. NDM-1 was first
Wellcome Trust Center, Kilifi,
Kenya (S Kariuki PhD); Division can be rapid and have severe consequences for vulnerable detected in South Africa in September 2011,20 and of
of Clinical Sciences, St George’s, hosts. Carbapenem resistance among common Entero- 70 carbapenem resistant enterobacteria isolates from
University of London, London, bacteriaceae has increased sharply over the past decade. In private and public hospitals received by the Antimicrobial
UK (Prof A Coates MD);
2012, 4·6% of acute-care hospitals in the USA reported at Resistance Reference Laboratory between May and July
European Association of
Pharmaceutical Industries and least one health-care associated infection caused by 2013, 19 tested positive for NDM-1.21
Associations, Brussels, Belgium carbapenem-resistant enterobacteria. The proportion of In India, E coli isolated from urine cultures of pregnant
(R Bergstrom MSc); Enterobacteriaceae that were resistant to carbapenems women in their first trimesters in the community (n=1815)
Michael G DeGroote Institute
increased from 0% in 2001 to 1·4% in 2010, with most of showed highest overall resistance to ampicillin, naladixic
for Infectious Disease Research
and the Department of the increase recorded in Klebsiella spp.1 acid, and co-trimoxazole, as 75%, 73%, and 59%,
Biochemistry and Biomedical Health-care associated infections are also increasingly respectively, between 2004 and 2007.22 30% showed
Sciences, McMaster University, recognised in LMICs. Findings of a recent review12 showed resistance to injectable antibiotics, such as amino-
Ontario, Canada
that pooled prevalence of health-care associated infections glycosides (represented by gentamicin). In a study of
(G D Wright PhD,
E D Brown PhD); and ReAct, in resource-limited settings (15·5 per 100 patients) was bloodstream infections,23 the proportion of E coli producing
Department of Medical twice the average prevalence in Europe (7·1 per ESBLs increased from 40% in 2002 to 61% in 2009, and the
Sciences, Uppsala University, 100 patients). Incidence of infections acquired in intensive proportion of K pneumoniae with carbapenem resistance
Uppsala, Sweden
(Prof Otto Cars MD)
care units in developing countries (pooled density 47·9 per increased from 2·4% to 52%.
1000 patient-days) was three times the rate in the USA Increasing rates of resistance to colistin and polymyxin B
Correspondence to:
Prof Otto Cars, Swedish Institute (13·6 per 1000 patient-days). Health-care associated infec- in Gram-negative organisms are being reported from
for Communicable Disease tions in neonatal intensive care units in some countries countries around the world, including South Korea,24 Italy,25
Control, 17182 Solna, Sweden (15·2–62·0 infections per 1000 patient-days) are up to nine Greece,26,27 and Saudi Arabia.28 Moreover, there is some evi-
[email protected]
times more common than in the USA (6·9 infections per dence of cross-resistance to colistin and host antimicrobial
1000 patient-days). Both the need for antibiotics and the peptides that are part of the body’s immune response.29
burden of resistance are likely to increase with the rate of Hospital-acquired MRSA arises worldwide. In high-
health-care associated infections in LMICs. income countries, it is being tackled with a combination of
These trends are globally consistent. Hospital data from new antibiotics and better hospital infection control, but
developing countries suggest that resistance to the WHO- community strains of MRSA continue to proliferate.30 In
recommended regimen of ampicillin and gentamicin in LMICs such as South Africa, 52% of 1147 S aureus viable
pathogens causing neonatal infections (in the first 28 days isolates from hospitalised bacteraemic intensive care unit
of life) is common: 71% of isolates of Klebsiella spp and 50% patients were MRSAs. Gram-positive infections are less
of E coli are resistant to gentamicin.13 Resistance is also a common in India, but high rates of MRSA in clinical
problem in early-onset, presumably maternally acquired, isolates in various studies in India have been documented
neonatal infections reported from hospital series in as 54·8% (range 32–80%).31 In Pakistan, rates of MRSA
developing countries. 60–70% of E coli and nearly 100% of have been fairly consistent since the mid-2000s at roughly
isolates of Klebsiella spp are ampicillin resistant, and 50%.32 However, community-acquired MRSA are
40–60% are resistant to gentamicin.14 High rates of ESBL increasingly reported, and rates range from 5–10%.33
production in E coli have restricted the use of second-line
treatment with extended-spectrum cephalosporins.15 Many The high burden of resistant infections
newborn babies in hospitals in south Asia are now treated Although poorly quantified, the global burden of resistance
with carbapenems as first-line treatment for sepsis or is probably concentrated in three major categories: longer

1058 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

duration of illness and higher rates of mortality in patients sepsis can die (vs a quarter of neonates with non-ESBL
with resistant infections, increasing costs of treatment for infections), and a half of neonates with MRSA die (vs 21%
resistant infections, and inability to do procedures that rely of neonates with MSSA).40 At these rates of mortality,
on effective antibiotics to prevent infection. roughly 106 514 neonatal deaths are attributable to Gram-
An estimated 25 000 people die every year in Europe negative organisms and S aureus, and 58 319 are attributable
from antibiotic-resistant bacteria.34 In the USA in 2005, an to ESBL resistance and MRSA in India alone.
estimated 94 000 invasive MRSA infections required In addition to the costs to human lives, high economic
hospitalisation and were associated with 19 000 deaths.35 A costs for health care exist,41 and these resources could be
recent report by the US Centers for Disease Control and deployed elsewhere.42 Resistant infections are more
Prevention conservatively estimated that at least 2 million expensive to treat and patients infected with resistant
illnesses and 23 000 deaths a year in the USA were caused strains of bacteria are more likely to require longer
by antibiotic resistance.36 These estimates are useful for hospitalisation and face higher treatment costs than are
suggesting scale, but imprecise because resistant infec- patients infected with drug-susceptible strains.43,44
tions are more common in individuals on long courses of According to one estimate, between 1997 and 1998,
antibiotic treatment; it is difficult to ascertain whether increases in drug resistance raised the cost of treating ear
resistance is the cause of death or a correlate of long infections by about 20% (US$216 million).45 Reduced
antibiotic treatment, hospitalisation, and underlying Streptococcus pneumonia sensitivity to penicillin in many
sickness. Few reliable estimates are available for LMICs, parts of the world has resulted in the need for more
but the higher burden of infectious disease and restricted expensive antibiotics including fluoroquinolones, oral
access to new antibiotics suggest a higher burden than in cephalosporins, and macrolides, driving up the cost of
high-income countries. treatment. In vitro resistance of S pneumoniae to β-lactams,
Findings of a study37 of patients with bloodstream macrolides, and other antibiotics has increased worldwide
infections and pneumonia in 537 intensive care units in as a result of the global dissemination of a few pandemic
ten countries showed that the risk of death (hazard ratio) clones. However, the most of roughly 826 000 pneumo-
associated with antimicrobial resistance (additional to that coccal disease deaths in children younger than 5 years
of the infection) was 1·2 (1·1–1·4) for pneumonia and likely result from poor access to antibiotics rather than
1·2 (0·9–1·5) for bloodstream infections caused by bacteria drug resistance.46 Although linezolid remains active
resistant to ceftazidime (A baumannii or Pseudomonas against most Staphylococcus spp, resistance has emerged
aeruginosa), third-generation cephalosporins (E coli), and and is moving upwards.47 Linezolid resistance has also
oxacillin (S aureus). Attributable mortality risk was highest been reported in Enterococcus faecium48 and Enterococcus
for S aureus in both pneumonia and bloodstream faecalis.49
infections. However, antimicrobial resistance did not A third consequence of resistance is related to the
significantly increase length of stay. P aeruginosa caused inability to do other interventions such as surgery, trans-
the highest burden of health-care-associated infections plantation, and chemotherapy.50 Investigators of a recent
because of its high prevalence and the pathogenicity of study estimated that, without effective antibiotics, 30–40%
both its drug-sensitive and drug-resistant strains. of patients having total hip replacements would have a
Multistate models, used to address the temporal postoperative infection, with a case-fatality rate of roughly
dynamics of admission, infection, discharge, and death, 30%.51 This burden, although poorly quantified, affects all
have found that, compared with patients without S aureus countries and is likely to be the main way in which
bacteraemia, the death hazard was 5·6 times greater with resistance drives up health-care costs.
MRSA (95% CI 3·36–9·41) and 2·7 times greater with
meticillin-sensitive S aureus (MSSA) bacteraemia (95% CI Uses of human and agricultural antibiotics are increasing
1·33–5·39).38 After adjustment for comorbidity, Antibiotic use is a main driver of selection pressure that
hospitalisation, age, and sex, the death hazard was contributes to resistance, and because consumers do not
2·9 times greater with MRSA (95% CI 1·70–4·88) and understand this problem, the drugs are among the world’s
1·7 times greater with MSSA (95% CI 0·84-3·47). A long- most commonly purchased. Most antibiotics are used
term follow-up study39 of 2000 patients with S aureus or unnecessarily, in commercially driven agriculture, and by
E coli bloodstream infections found an 80–150% increase physicians uncertain of a diagnosis or treating largely self-
in mortality associated with resistance at 30 days after limiting bacterial or viral infections. In high-income
infection. At 90 days, MRSA had twice the attributable risk countries, patients with resistant infections can turn to
of death relative to MSSA.38 more expensive, newer-generation antibiotics, but in
In LMICs, where the ability to pay for second-line drugs developing countries, where infectious diseases are
is limited, worse health outcomes, especially in neonates common and the burden is high, patients might be unable
are more common. Even with effective antibiotics, to obtain or to afford second-line treatments.
infections are the major cause of neonatal deaths, which in Large differences in the frequency of resistant
turn account for more than a third of the global burden of infections have been noted, both across European
child mortality.13 More than half of neonates with ESBL countries52 and among regions of the USA.53,54 Variations

www.thelancet.com/infection Vol 13 December 2013 1059

 The Lancet Infectious Diseases Commission

in antibiotic consumption, both between and within In Japan and the USA, patients drive expectations for
countries,52,55–59 might explain the differences. Con- antibiotic prescribing. In China, hospitals that rely on
sumption of antibiotics in countries reporting data to pharmaceutical sales for income have an incentive to
the European Surveillance of Antimicrobial Con- overprescribe; one study62 estimated that a quarter of
sumption Network (ESAC-Net) in 2010 varied from revenue in two hospitals came from antibiotic sales. In
11·1 (Estonia) to 39·4 (Greece) defined daily doses per India, doctors routinely receive compensation from drug
1000 inhabitants per day.60 Consumption of carbapenems sellers in exchange for directing patients to their
increased in 15 of 19 countries reporting data to ESAC- pharmacies. Antibiotic sales increase with insurance
For more on ResistanceMap see Net between 2007 and 2010. Data from ResistanceMap status of patients because patients with insurance are
http://www.cddep.org/ from the USA suggested that between 2007 and 2010, likely to be less price sensitive.63 Antibiotic prescribing
ResistanceMap/key-findings#.
UnJxKfk73To
there was a downward national trend in outpatient might also be affected by competition between health-
antibiotic consumption. Prescriptions fell by 17% care providers; in Taiwan, a one standard deviation
between 1999 and 2010. However, states in southeast increase in competition raises antibiotic prescription by
USA continued to consume more than twice the amount up to 2·4%.64 Competition from unsanctioned providers
of antibiotics per person than did those in the Pacific also exacerbates competitive pressure on legitimate
northwest and New England. medical professionals. Little evidence exists that trained
Worldwide, antibiotic consumption is on the rise providers give more appropriate treatment than do
(figure 1). Although carbapenems are expensive, sales in untrained pharmacy attendants, perhaps because
Egypt, India, and Pakistan have increased with over-the- pharmacists often mimic prescription patterns of other
counter availability. local providers and unwittingly copy both desirable and
Non-prescription antibiotic use is common in many undesirable practices. A study65 from Thailand found that
LMICs, where ensuring that people who truly need a pharmacy’s proximity to a hospital improved the
antibiotics have access while discouraging unnecessary appropriateness of antibiotics sold.
use is a challenge. Non-prescription use accounts for Many drivers of antibiotic consumption are based in
19–100% of antibiotic use outside northern Europe and human medicine, but antibiotic use in veterinary medicine
North America.61 Even when prescriptions are needed to and for growth promotion and disease prevention in
obtain antibiotics, physicians might not adequately screen agriculture, aquaculture, and horticulture is also a major
for appropriate use. contributing factor. Although precise estimates are scarce,

1·8
Standard units per 1 000 000 population

1·2

0·6

0
20 05
20 6
20 07
2008
20 9
10

20 5
20 6
20 07
2008
20 9
10

20 05
20 6
20 07
2008
20 9
10

20 05
20 6
20 07
2008
20 9
10

20 5
20 6
20 07
2008
20 9
10

20 5
20 6
20 07
2008
20 9
10

20 5
20 6
20 07
2008
20 9
10

20 05
20 6
20 07
2008
20 9
10

20 05
20 6
20 07
2008
20 9
10
0

0
0

0
0

0
20

20

20

20

20

20

20

20

20

Netherlands USA French West Africa Brazil Vietnam Indonesia India Pakistan Egypt

Figure 1: Trends in retail sales of carbapenem antibiotics for Gram-negative bacteria

Based on data obtained from IMS Health’s MIDAS™ database. *An IMS grouping of Benin, Burkina Faso, Cameroon, Congo (Brazzaville), Gabon, Guinea, Ivory Coast,
Mali, Senegal, and Togo.

1060 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

of the crudely estimated 100 000–200 000 tonnes of

antibiotics manufactured every year,66 most goes to the
agricultural, horticultural, and veterinary sectors. The
finding of low levels of resistance in polar bears on the
isolated Arctic archipelago of Svalbard supports the
hypothesis that ecological resistance close to human
settlements is anthropogenic in origin (figure 2).67
Although the transfer of antibiotic resistance plasmids
from treated animals to human beings has been long
suspected,68 findings from recent studies using whole-

Louise Murray/Science Photo Library

genome sequencing have confirmed animal-to-human
transfers of resistance genes.69
A global system for surveillance of antibiotic use and
resistance and its health and economic burden is urgently
needed. Surveillance should include environmental
sampling in addition to examination of clinical isolates.
Figure 2: Antibiotic resistance in nature, such as that identified in polar bears in Svalbard, is likely anthropogenic
Part 2: Getting out of the impasse
How did we end up here?
The lack of understanding of the unique features and Moving on
risk of resistance has paved the way for the present National approaches and commitment
epidemic. Moreover, few studies have been done on the Countries that have implemented comprehensive
magnitude of the burden to convince policy makers of national strategies have been the most successful in
the urgent need to react. Since the penicillin era, controlling resistance.76–79 These strategies include, but
antibiotics have been viewed as wonder drugs that could are not restricted to, good health-care infrastructure and
be prescribed without fear of harm, despite early health insurance for all; limited drug advertising;
warnings of consequences such as antibiotic resistance surveillance of antibiotic use and to detect resistance in
and side-effects.70,71 Their use has spread into many non- human beings and animals; policies for prudent
medical areas, and has been unregulated, both legally antibiotic use in human beings and animals; standardised
and illegally. Antibiotic resistance is perceived as a infection control policies and sufficient staffing; anti-
complex medical problem. Antibiotics are different from biotic stewardship programmes in hospitals and other
all other drug groups in that the effects of their use health-care facilities; and isolation or decontamination of
extend far beyond individual patients. Even more patients with resistant organisms.76 In particular, several
worrying is the accumulating evidence that antibiotic use European countries have introduced these strategies.
in seriously ill but uninfected patients can actually Additionally, countries with cases of antibiotic resistance
increase mortality.72 have found a targeted national approach successful—eg,
Traffic and selling of antibiotics at markets, shops, and the UK for control of MRSA and Clostridium difficile.77
pharmacies is largely unregulated, without prescription, Israel controlled KPC by a national approach,78 and the
and even without involvement of a person with USA has implemented various initiatives.79 However,
pharmaceutical training.61 This widespread access is made these programmes need time and patience to be set up
easier by the internet and marketing stunts for free or and need to be backed by visionary governments with
cheap antibiotics. Selective information and material adequate funding. A stepwise approach to a national
incentives from pharmaceutical companies influence strategy according to a contextualised and prioritised
doctors in affluent countries, leading to some regulation of road map might be the best way forward for most
their activities; however, with new markets emerging, settings. In resource-poor countries, there has been
more influence is to be expected in low-resource settings, much less progress, although China and India notably
which often do not have other sources of information and have made important steps recently. A meeting of
training. For instance, prescribers in Kisangani, the professional societies in India issued its Chennai
Democratic Republic of Congo, mentioned pharmaceutical declaration80 and the Chinese Government has enacted
companies as the first source of information about policies to restrict antibiotic use, including the initiation
antibiotics.73 Production and use continues to increase of a campaign on antibiotic resistance, stratification of
generally in an uncontrolled way in developing countries.74 antibiotic use, and enforced restriction on drug pre-
In the non-medical arenas of agriculture, aquaculture, scription.81 Proposals for the stepwise development of For more on the Global
and intensive farming, huge amounts of antibiotics are activities of a national task force based on experiences Antibiotic Resistance
Partnership see http://www.
used in some countries—up to four-times the amount from several LMICs are being developed by the cddep.org/projects/global_
used in human medicine.75 There is little separation of the Global Antibiotic Resistance Partnership (Ramanan antibiotic_resistance_
types of antibiotic used in human beings and animals. Laxminarayan, Personal Communication). partnership

www.thelancet.com/infection Vol 13 December 2013 1061

 The Lancet Infectious Diseases Commission

Rational antibiotic use in hospitals who learn how to use antibiotics from their prescriptions.2
The set of activities and policies to improve the rational use Apart from medical training, physicians are influenced by
of antibiotics is also known as antibiotic stewardship. their peers, and perceived demands of patients. Therefore,
Essential elements of an antibiotic policy include a stable physicians might find it difficult to comply with treatment
and restrictive list of antibiotics in use, standard treatment guidelines.90 These barriers to compliance should be
guidelines, audit and feedback of prescriptions, removed or minimised, and options for alternative
surveillance of bacterial resistance and antibiotic use, and actions for guideline compliance should be simul-
education at all levels.82 Antibiotic stewardship has typically taneously provided.91 Examples of options for non-
been developed in the hospital context in resource-rich antibiotic treatment in viral or self-limiting infections are
countries, but stewardship activities should be expanded to the prescription of herbal medicines, as opposed to
primary care on a national level. Its combined goals are antibiotics,92 and use of a delayed prescription technique
improved outcomes for patients, containment of antibiotic with explicit instructions for patients about when to use
resistance, and increased cost-effectiveness of care. antibiotics.93
Antibiotic stewardship should be done by all health-care To encourage guideline compliance, consequences of
facilities and should be part of accreditation programmes. irrational use of antibiotics should be reframed to be
Effective stewardship programmes can decrease relevant to the self-interest of prescribers and
antibiotic use by 20–40%, incidence of health-care- institutions. Motivational measures include pay-for-
associated infections (C difficile, MRSA, and others), performance policy,92 the audit-feedback mechanism on
lengths of stay, and prevalence of bacterial resistance.83,84 antibiotic prescribing rates of individual prescribers,94
Ideally, stewardship teams should include an infectious and public disclosure on antibiotic prescribing rates of
diseases physician, a (clinical) pharmacist with infectious each health-care facility or area.95 Major challenges arise
diseases training, a clinical microbiologist, information when antibiotic prescriptions are a source of revenue
system specialist, infection control professional, and a for individuals or institutions, either by a fee-for-service
hospital epidemiologist, but such specialists might not be remuneration scheme96 or drug-promotion incentives.97
available.85 In this case, a combination of the audit-feedback and
Barriers to the implementation of effective and the public disclosure measures seems to be useful.
sustainable programmes exist in many regions of the However, in many resource-poor countries, doctors
world. The bottlenecks for implementing stewardship in own the pharmacies and, therefore, the financial
both resource rich and poor countries are often strikingly incentives to prescribe expensive antibiotics are even
similar, largely as a result of insufficient leadership, greater than with a fee-for-service reimbursement. In
commitment, and funding. Standard treatment guidelines, these settings, separation of prescription and
which are essential to steer the use of antibiotics, can be dispensing activities are needed.
counterproductive if not updated regularly and well On the demand side, self-medication by consumers
implemented. National guidelines, particularly in LMICs, with antibiotics purchased without a prescription is
if present, are often outdated or not disseminated to the common, especially in eastern and southern Europe,
genuine prescribers, who in turn might use a combination Africa, South America, and Asia.61 Consumers have
of outdated textbooks, international but locally irrelevant positive attitudes towards antibiotics, but paradoxically
guidelines, or just the habits of their teacher.86 Concentrated they have poor knowledge about these drugs and
efforts at national levels are needed to invest in the revision diseases.98 The availability of antibiotics without
of standard treatment guidelines into clear, simple, prescription—an important enabling factor—mainly
updated, evidence-based, locally relevant, and accessible results from absence of prescription-only regulation,
documents. A recent national project launched in Vietnam ineffective law enforcement, poverty-driven practice,
has, as one of its objectives, the development of updated culture, and norms.99,100 Suboptimum compliance on use,
guidelines for several important infectious disease including taking leftover antibiotics from previous
syndromes.87 Establishing the resistance threshold for treatment courses and sharing unused drugs with other
particular infections warranting the switch to an alternative people, is common in both developed and developing
empirical antibiotic is a particular challenge.88 countries.101 Improvements in health-care financing
might positively affect access to a full course of antibiotics.
Rational antibiotic use in the community Ideally, consumers should have access to accurate
A programme on rational antibiotic use or antibiotic information on antibiotics and infectious diseases
stewardship in the community should cover a wide range instead of access to antibiotics without prescription. A
of settings, such as ambulatory care facilities, pharmacies, ban of over-the-counter sale of antibiotics has been
drug vendor outlets, households, and agriculture. implemented in many countries including Chile.102
Overuse and irrational use of antibiotics, either driven by Some countries launch national campaigns (eg, the US
For more on the Get Smart the supply or demand sides, have been documented in all Centers for Diseases Control and Prevention’s Get Smart
campaign see http://www.cdc. these settings.89 On the supply side, physicians are often or Antibiotic Awareness Day in the European Union
gov/getsmart/
role models for other health professionals and patients [EU]), with the aims of improving knowledge of

1062 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

resistance and lowering of antibiotic use among Infection control

consumers and prescribers.103 Better than treatment of infection is, of course, prevention.
For countries with inadequate health delivery systems, From a resistance perspective, prevention reduces anti-
prescription-only regulation might impede access to biotic use and the spread of resistant bacteria; however,
antibiotics. Although regulation is crucial to safeguard prevention is not the main strategy to control resistance
access to antibiotics, a transition towards such regulation because antibiotic use also needs to be controlled.
needs governmental commitment and improvements in Nevertheless, at the community level, improvement of
health systems that are not possible in many countries. sanitation, access to clean water, poverty reduction, and
Hence, antibiotic stewardship programmes need to be vaccination will have a huge effect on both infectious
adjusted to local conditions—eg, use of telemedicine to disease incidence and transfer of and colonisation with
help with treatment decisions and the referral process104 resistant genes and multidrug-resistant organisms.
and to help community pharmacists make good decisions At the hospital level, prevention of health-care-associated
about antibiotic dispensing. More research is needed on infections, which are often multidrug resistant, is essential,
how to balance the effects of specific interventions on but challenging. Besides hand hygiene, the importance of
individual health versus increasing the resistant bacterial which cannot be overemphasised, benchmarking (open
population. In low-resources countries, stewardship comparison of between health-care facilities) of frequencies
programmes to empower drug vendors on rational use of of health-care-associated infections is useful to decrease the
antibiotics seem promising, but further research is number of these infections. Observation makes one more
needed.105 careful; most of the infection control policies in place have
been developed around MRSA, vancomycin-resistant
Education and changing social norms enterococci, C difficile, catheter-related bloodstream
When irrational use of antibiotics repeatedly happens infections, catheter-associated urinary tract infections, and
among the public and health professionals, it becomes the ventilator pneumonia. Infection control interventions need
norm. To break this pattern, antibiotic stewardship to be reassessed and improved in an era with multidrug-
programmes should focus not only on appropriate use, but resistant Gram-negative bacilli and mobile antibiotic
also on ensuring sustainability of behavioural change and resistance genes. Additionally, cultural barriers for
reorientation of social norms.92,106 Such a stewardship implementation of basic hygiene procedures are probably a
programme in an LMIC is the Antibiotic Smart Use widespread problem and need much more study.
Program in Thailand.92 Many bottlenecks remain in the
promotion and sustainability of good prescription Role of diagnostics
practices, especially with regards to social norms. Solutions Efforts to improve microbiological laboratories are
need to focus on multifaceted and multilevel interventions underway. Speed of testing and laboratory automation
that define local barriers and beliefs, which can vary widely have been the focus of recent developments. In trying to
between cultures, countries, and regions. reduce testing time, various methods have been developed
Education of all health-care workers, laboratory staff, (eg, PCR-based tests, various point-of-care tests, and
veterinarians, and the public on appropriate antibiotic use MALDI-TOF mass spectrometry). These tests are
and antibiotic resistance is essential, and educational supplementary to detect disease-causing pathogens next to
strategies have recently been reviewed.107 Although traditional culture-based methods because they detect only
education alone might not be powerful enough as an the pathogens for which the test is designed.110,111 Diagnostic
intervention, it generates knowledge that is essential for methods that reduce antibiotic use or narrow the spectrum
health-care workers to understand and support the should be promoted. Even if these methods do not have a
resistance control programmes; such education should be direct benefit on clinical outcome, reduction of antibiotic
started very early in the medical curriculum.107 The role of pressure can help to slow induction and spread of
up-to-date undergraduate and postgraduate education is resistance.112–114
even more important in settings with restricted access to Besides direct care of patients, the results of diagnostic
medical literature. microbiology testing are used to inform local, regional,
Clear information for policy makers about antibiotic and national surveillance systems. A successful example
resistance and its effect on public health has a crucial role is the significant reduction in MRSA bacteraemia in
in making this complex problem tangible; initiatives such hospitals in the UK since implementation of mandatory
as the Drug Resistance Index might help to achieve this.108 MRSA surveillance in 2001.115 Surveillance of bacterial
Educational and awareness campaigns for the general resistance generates essential information, which
public might help to generate an understanding that can promotes and directs stewardship activities. The scarcity
support the prescriber to withhold antibiotics.109 Although of quality-assured microbiology laboratories in low-
hard endpoints are difficult to define and to measure, resource settings and lack of priority given in the past
these campaigns seem to contribute to more careful use of decades to sustained bacterial surveillance have led to
antibiotics. Education should be tailored and started early large empty areas on the worldwide resistance maps,
on to shape behaviour rather than having to change it. especially for sub-Saharan Africa and rural Asia.116

www.thelancet.com/infection Vol 13 December 2013 1063

 The Lancet Infectious Diseases Commission

Research
Research efforts need to be focused on intervention
strategies and solutions rather than doom-and-gloom
reports. Until recently, MRSA and vancomycin-resistant
enterococci governed the resistance agenda, but now,
multidrug-resistant Gram-negative bacteria are the main
cause for concern. However, findings (summarised in the
recent Cochrane reports122 and elsewhere) show that
community and hospital antibiotic stewardship
interventions can modulate prescription enough to reduce
Peter Chadwick/Science Photo Library

resistance in many organisms in some settings. Time-

series analysis can be used by all hospitals with
computerised databases of antibiotic consumption and
resistance rates. This approach allows interventions to be
tailored to individual scenarios and prediction of resistance
rates, and this information can be used to develop local
Figure 3: Waste-water treatment facilities can be hotspots for horizontal transfer of resistance
treatment guidelines.
Besides antibiotics, new treatment strategies under
investigation include methods to stop plasmid replication123
Quality-assured basic microbiology services and routine or resistance mechanisms such as efflux pump inhibitors.124
bacterial resistance surveillance are urgently needed in Furthermore, bacteriophage treatment—used in the 1920s
most of these areas. Positive experiences in high-income and later in the Soviet era—is being investigated as another
settings, such as Europe,117 with its easy-to-understand, potential strategy, but regulatory requirements for these
interactive, and yearly updated data, could inspire policy types of drugs are challenging,125 and their use might not
makers in other regions. The addition of antibiotic extend to life-threatening infections.
resistance to death registers might help to raise
awareness of antibiotic resistance on the priority agenda Safeguarding the future
of policy makers. Will any set of interventions be effective enough, in view of
the present wave of antibiotic resistance? Suggested
Beyond use in human beings interventions are not optional, they are basic requirements
Bolder interventions outside hospitals and a move to to ensure rational use of antibiotics and optimum outcomes
ecological antibiotic stewardship are needed. Strategies for patients. Enlightened national and global leadership is
should be focused on control of non-human sources of needed along with sufficient technical capacity at all
antibiotics, resistant bacteria, and resistance genes, such different policy levels. Comprehensive national and
as agriculture and waste water from the pharmaceutical international plans, like those in the EU,126 are needed, and
industry. The issue of antibiotic resistance should be part should have similar visibility and effect to those for other
of the one-health movement. Use of antibiotics as growth important health problems such as HIV, tuberculosis, and
promoters should be banned worldwide as has happened malaria. Additionally, these national plans need to take into
in the EU. In Denmark, the banning of growth promoters account their interfaces with health-care organisation,
in livestock resulted in a net reduction of animal quality assurance and financing, and professional
antibiotic use with no or minor increases in production education. The scientific community should clarify the
costs without increased incidence of zoonotic infections causes, scale, and fast pace of the evolution of resistance.
in human beings.118 Worryingly, last resort drugs, such as Antibiotic resistance should be on the global political
colistin, are being used extensively in agriculture.119 agenda, not just the agendas of infectious disease meetings.
Recent initiatives, also in the EU, are trying to limit the Emphasis is placed on reinvigoration of the drug
use of this drug in agriculture. discovery industry, but there is a sense that all low-hanging
The environment is key in the spread of resistance. For fruit have already been picked and that new developments,
example, wastewater treatment facilities can be a hotspot even if successful, cannot sate demand and will be only a
for horizontal resistance gene transfer (figure 3).120 temporary fix. Many people believe that with billions of
Strategies to reduce this mode of transfer are warranted, years of evolution, bacteria will always be better genetic
including neutralisation of antibiotics in wastewater and engineers than people. Antibiotics are a natural product of
in the environment generally. Findings from a recent bacteria and so resistance mechanisms are not new.
study121 showed that chlorination of drinking water can Therefore, antibiotics are a precious public good and their
actually concentrate some antibiotic resistant genes. intended and unintended environmental release needs to
Research efforts need to focus on how to reduce and be monitored and controlled.
neutralise manmade antibiotic pressure and how to Stewardship efforts might win a battle, they will
control the resistance gene pool in hotspot environments. certainly not win the war. Antibiotic stewardship—an

1064 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

integral part of standard care provision—should be part the potential for changing the landscape of diagnostic
of hospital accreditation. Besides, a holistic, ecological, testing became clear with the development of the PCR in
one-health approach is needed. There is an urgent need the 1980s.133 Since then, we have seen a technological
to stop enriching the resistance gene pool with revolution with the development of many complex, highly
unnecessary antibiotic pollution. The issues of antibiotic specific molecular diagnostic assays. These systems can
resistance are akin to those of global warming due to decrease the time needed for detection of biomolecules,
excess use of carbon-based fuels with the resultant like proteins and nucleic acids, from a few hours to a few
pollution, and we need a worldwide agenda, perhaps like minutes, and should greatly improve medical diagnostics.
the Kyoto agreement, but with much more adherence.127 However, most of these technologies have not yet reached
Solutions will not be easy and, paradoxically, might clinical diagnostic laboratories.
include increasing the price of antibiotics to put a true
value on their use, while maintaining the delicate balance Defining the medical needs
between overuse versus lack of access. There is room for Many available rapid diagnostic tests are designed on the
innovative ideas in quality assurance, health financing, basis of microbiological grounds and detect as many
and social marketing. microbes as possible, rather than on clinical grounds to
The future of antibiotics and survival of every human address the real medical need. Furthermore, no consensus
being that acquires a bacterial infection will depend on the exists on how quickly tests should produce results to
serious commitment of many stakeholders, including identify patients who really need an antibiotic—should
government authorities, policy makers, health-care companies only invest in developing technologies that
workers, university teachers, pharmaceutical companies, produce results in less than 30 min in primary care or less
and consumers. than 1 h in hospital care? If so, very few companies have
technologies in their pipelines that can meet these
Part 3: Minimising the time to effective requirements. So should a first dose of antibiotics be given
treatment—rapid diagnostic testing and then treatment adjusted on the basis of diagnostic test
Diagnostic uncertainty drives irrational use results?
Diagnostic (viral or bacterial cause) or prognostic (life- How do the needs (eg, speed, robustness of system, cost,
threatening or self-limiting infection) uncertainty makes it and user friendliness) in industrialised countries compare
difficult for clinicians to know when to provide and when with those in lower resource settings? And how will these
to withhold antibiotic treatment. Consequently, antibiotics needs fit with the different health-care models and
are overused in hospitals and outpatient settings, resulting reimbursement systems? Should companies develop tests
in increased antibiotic resistance52 and the pandemic that identify pathogens and detect resistance, or is the
spread of highly resistant bacterial clones.128 Findings of identification of which organisms caused the infection not
studies of patients with acute cough—one of the most important as long as doctors know which antibiotics are
common reasons for consultation in primary care and needed? Which antibiotic resistance genes are always
antibiotic prescription in high income countries—showed expressed in vivo and could therefore be targeted in the
that antibiotics did not meaningfully change important system? Quantitative microbial cultures with cut-offs
outcomes.129,130 Since the early days of discovery of bacteria, (expressed in colony forming units per mL) are used to
culture-based assays have remained the gold standard for distinguish between colonisation and infection, but should
identification of pathogens and susceptibility testing. technologies be developed that also define molecular load
However, these methods are slow, typically identifying thresholds (expressed in number of DNA copies per mL)
causative pathogens in at best 24 h, and returning to distinguish the colonisation and infection status of
susceptibility results in 48 h. Minimisation of time to potential pathogens in different types of samples? How
effective treatment decreases morbidity and mortality in should diagnostics be used in clinical trials to identify
severe infection.131 patients infected with targeted pathogens or multidrug-
The effects of antibiotic resistance on human health are resistant organisms? Many companies are struggling to
probably highest in countries with the lowest income align their business goals with the technology solutions
because the spread of resistant bacteria is facilitated by because these fundamental questions have not been
poor hygiene, contaminated food, polluted water, properly addressed by experts in the specialty. A technology
overcrowding, and increased susceptibility to infection road map on rapid diagnostic tests for infectious diseases
because of malnutrition or HIV. Personalised medicine is needed to help forecast, plan, and coordinate technology
based on novel and rapid diagnostic strategies should help developments that meet real medical needs.
identify patients who need antibiotics. In many such
settings, the need for alternative technologies is also Huge technical challenges
pressing, because routine culture and susceptibility testing Molecular tests can reduce the time to yield results but
are not provided, even to support diagnosis of life- come with many drawbacks, including complex sample
threatening infections like pneumonia and meningitis preparation, little integration of the different steps,
(figure 4).132 Immunoassays provided a faster option, but inability to handle large volumes or multiplexing for

www.thelancet.com/infection Vol 13 December 2013 1065

 The Lancet Infectious Diseases Commission

impedance and magnetic resonance) and several national

and global initiatives are supporting the development of
these new technologies.

Lack of guidance for assessments

Few guidelines for the assessment of clinical diagnostic
tests for infectious diseases exist. Companies often claim
sensitivities and specificities close to 100% without
oversight on the design and conduct of the diagnostic
assessments. Moreover, these values are calculated with
spiked or archived samples, and are not indicative of the
real-world situations in which tests will be used. Other
factors often not taken into account (eg, conditions of
Biophoto Associates/Science Photo Library

storage and shelf life) are also important, especially in

LMICs.135 If standards are available for the assessment of
diagnostic tests, they are published in industrialised
countries and are not necessarily applicable to diseases
prevalent in other settings.136 Finally, few studies
investigated rapid diagnostic tests in terms of effect on
antibiotic use or resistance, or patients’ outcomes.
Figure 4: Culture-based methods remain the cornerstone of diagnosis and resistance testing
Poor performance of molecular tests
Despite excellent analytical sensitivity and specificity, the
detection of many different targets, and high cost. Ideally, test might perform poorly in clinical trials or routine use,
so-called sample-in to results-out technologies are needed depending on study groups (screening vs diagnosis,
that integrate sample preparation, amplification, symptomatic vs asymptomatic, active vs latent infection),
detection, and analysis. The system should be able to setting (low vs high prevalence), complexity of test (done by
detect pathogens and host biomarkers (proteins and trained vs unskilled staff), and comparator (more or less
nucleic acids) simultaneously. Sample preparation sensitive than the comparator test). Assessment of
remains the largest bottleneck in miniaturised molecular tests has been mainly analytical and few were
diagnostics:134 large volumes (several millilitres of blood) extensively assessed with clinical specimens from clearly
need to be reduced to small amounts (in the order of defined populations of patients from a wide geographic
microlitres); the microbial load in the sample can vary a area. Findings from multicentre studies showed significant
lot and be very low; the target should remain intact, but variations of detection rates of molecular tests between
many complex specimens contain nucleases or inhibitors laboratories, with different or even the same tests, even
of nucleic acid amplifications. Therefore, most sample though some of the laboratory personnel were very
preparations of available miniaturised molecular systems experienced with the use of amplification assays.137,138
rely on many operations, and need several liquid additions Standardisation of methods and reference reagents is
and washing steps. needed for complete quality assessment programmes,
Most microfluidics-based systems rely on conventional including proficiency panels to assess the performance of
benchtop sample preparation, thus restricting their use in molecular tests. A network of designated diagnostic
point-of-care tests with limited or no laboratory access. laboratories that can assess molecular tests, alongside
Conversion of these off-cartridge sample preparation steps routine diagnostic testing, would be a step forward.
to cartridge-based microfluidic systems will be challenging
because the microscale physical conditions (eg, the No proven clinical benefit
surface/volume ratio) can alter the assay conditions. Assessment of the quality of molecular tests should take
Additionally, the reliability and cost of lab-on-a-chip clinical usefulness into account. However, diagnostic tests
systems typically increase with a factor per added are sold and used without good evidence of effectiveness,
component, such as fluidic interfacing, valving, pumping, especially in developing countries, and the endpoints that
microscale mixing, and waste disposal. Although should be used in clinical trials have not been agreed upon.
microfluidics offer great promise in the area of rapid Very few well designed preclinical and clinical trials have
diagnostics, a microscale assay needs to be carefully been done to show that rapid diagnostic testing improves
designed, in which novel solutions should provide clinical outcome or reduces antibiotic resistance. Although
integrated functionalities in a minimum amount of unit more recently, clinical trials on the performance and
steps and in a minimum cartridge size. Finally, most rapid effectiveness of diagnostic tests were funded by the public
diagnostic tests are based on detection of nucleic acids, but sector (eg, the European Commission’s Framework
several other technologies are being pursued (eg, Programmes), most trials are still sponsored by the

1066 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

industry (many of these are small biotechnology template and hence, are heavily reliant on conventional
companies, which have limited resources and expertise to culture.141 Second, whether susceptibility tests are enough
do such trials). or whether laboratories should still seek ESBLs and
No studies have assessed the best integration of tests carbapenemases directly is unclear.142 Third, limited data
into health-care practice. The danger is that novel and suggest an association between minimum inhibitory
expensive technologies could become representative of concentrations (MICs) and outcomes for patients with
high-quality care with little evidence to support their use. Gram-negative infections.143 Hence phenotypic testing
Technological innovations that allow more personalised based on MIC values might guide antimicrobial treatment
medicine are likely to raise rather than lower health-care better than do genotypic tests that detect the resistance
costs. A McKinsey report139 estimated that four countries mechanism. If the susceptibility test results (susceptibility
(Austria, Portugal, Spain, and the USA) would spend more category or MIC, irrespective of resistance mechanism)
than 20% of their gross domestic product on health care, should guide treatment, molecular assays for detection of
and only five of 21 Organisation for Economic Cooperation ESBLs or carbapenemases might have little value for
and Development countries (Denmark, Italy, the management of infected patients. For infection control
Netherlands, Sweden, and the UK) would spend less than purposes, these assays could still be very useful in the
15% by 2040. The USA would allocate nearly 30% of its detection of carriers and prevention of transmission.
economic output to health care by 2040. Therefore, if new
technologies are to be successfully implemented, we will Many barriers for use
need to move away from the business-as-usual approach, Research needs to move beyond comparison of point-of-
to develop new and smarter pathways of care, and to show care test performance with laboratory tests to an agenda of
that they are cost-effective. understanding barriers and opportunities regarding
uptake into routine care. Despite findings from several
What samples should be used? studies showing that simple and user-friendly tests for
No consensus exists on the best sampling sites for detecting C-reactive protein144 and procalcitonin145 are
detection of pathogens in many infectious diseases. Good effective in achieving important reductions in antibiotic
studies comparing different specimens are scarce or prescribing, these tests are still not widely used. Many
results inconsistent. Most of these studies are based on barriers, including physicians’ attitudes towards diagnostic
conventional diagnostic microbiology and cannot be testing, approval by regulatory authorities, recom-
extrapolated to molecular diagnostics. For example, acute mendation by guidelines, social, ethical, economical, and
community-acquired respiratory tract infections are the political factors, affect the uptake of new diagnostic
commonest reason for the prescription of antibiotics. technologies and delivery into health systems. Many
However, we do not know what the best specimens are to countries, especially in low-resource settings, do not
detect many of the pathogens causing respiratory tract regulate in-vitro diagnostics or require submission of
infections, or how to distinguish between organisms clinical trial data.136 Input and help from behavioural
infecting the lower respiratory tract and those colonising sciences and social marketing are needed to address
the rhinopharynx (eg, nasal aspirate, nasopharyngeal barriers to acceptance of rapid diagnostic tests and to help
aspirate, nasal swab, nasopharyngeal swab, nasal wash, understand motivational factors that could help to
oropharyngeal swab, or sputum). Depending on the overcome hurdles to effective use of these tests in
organism, distinct differences are seen when comparing management of patients.
different specimens by use of conventional diagnostic
microbiology.140 However, with more sensitive molecular Guidelines de-emphasise diagnostic microbiology
methods, such as nucleic amplification systems, the Many guidelines do not recommend conventional
differences in recovery rates between the respiratory diagnostics to identify the pathogens or they recommend
specimens might be more subtle. treatment initiation within a short timeframe (eg, treatment
of community-acquired pneumonia), leading to excessive
Role in antibiotic stewardship is controversial empirical treatment with broad-spectrum antibiotics.146
Over the past decade, the prevalences of infections caused Additionally, in developing countries, where access to
by ESBL-producing and carbapenemase-producing Gram- diagnostic laboratories is limited, patients presenting with
negative bacteria have increased significantly.128 A rapid a particular syndrome are treated for all major causes.
assay that can detect infection with such organisms should Although generally cheap, this approach results in
improve outcomes for patients because delay in initiation inappropriate antibiotic treatment without syndromic
of effective antimicrobial treatment tends to be associated diagnosis of disease. Rapid diagnostic tests for detection of
with increased mortality.131 However, whether rapid causative agents or biomarkers at the point of care are
diagnostics for detection of genetic resistance markers are needed to allow prompt and specific targeting with a
useful to guide treatment remains controversial. First, the narrow-spectrum antibiotic. Rapid diagnostics would boost
available molecular assays are not truly rapid diagnostic development of narrow-spectrum antibiotics because
systems because many of them require bacterial DNA as a companion diagnostics are a prerequisite to the prescription

www.thelancet.com/infection Vol 13 December 2013 1067

 The Lancet Infectious Diseases Commission

of these drugs. Improved diagnostics would also reduce the The effect of low doses of antimicrobials for growth
cost of clinical trials by enabling focused enrolment of only promotion on antimicrobial resistance has been
those patients infected with target pathogens. documented for several substances.152 For example, use of
the glycopeptide avoparcin was associated with the
Part 4: The interface between people and animals selection of vancomycin-resistant enterococci;153 after its
Antibiotic use in animals withdrawal, the prevalence of resistance decreased.154
Use of antibiotics in animals and its potential effect on
human health has been a controversy for at least half a Veterinary use
century, presently fuelled by the crisis of resistance. Generally, the most common indications for antimicrobial
Predictably the debate is polarised. Results of scientific treatment or preventive use in terrestrial animals reared
studies have sometimes been conflicting, which is for food production are enteric and respiratory disorders in
confusing for readers unfamiliar with the context. young animals and mastitis in dairy cows.155 Untreated,
Although the discussion here is restricted to terrestrial these disorders affect animal welfare and productivity and
animals excluding honeybees, aquaculture is also can sometimes lead to substantial mortality. Therapeutic
important in the overall discussion. use can include individual animals, but can also be
treatment of a group of diseased animals by injection or
Use for growth promotion orally. Preventive use can be anything from targeted
The growth-promoting effect of low doses of antimicrobials interventions to control the spread of a diagnosed disease
was discovered in the late 1940s.147 Feeding subtherapeutic in a defined group of animals to routine treatment of all
doses of antimicrobials became an integral part of animals during specific periods of stress such as weaning,
intensive rearing of animals. Undoubtedly, these practices after transportation, or when combining new animals with
supported the intensification of modern food production a herd or mixing animals from different sources.
by facilitating early weaning, increased animal densities, Preventive use is mostly given via feed or water. With some
and cheap feed sources.148 Furthermore, suboptimum exceptions, the antimicrobial classes used are the same as
growth caused by unsanitary conditions is sometimes those used in human medicine. However, some newer
compensated with addition of antibiotics to feed.148 types of antimicrobials, such as carbapenems, oxa-
Worldwide, many substances have been or are used, some zolidinones, and glycylcyclines are not used for animals
of which are not used in human medicine (eg, reared for food.
flavophospholipol) and some from classes that are (eg, the Regulations and practices vary widely around the globe
macrolide tylosin). How these substances lead to increased and are probably affected by the economic and social
growth rate is unclear, but prevention of enteric diseases, context. In the EU, all antimicrobials for systemic use in
such as weaning diarrhoea, probably has an important animals reared for food production are on prescription
role.148 Low doses of tetracycline also reduces morbidity only. In other parts of the world, antimicrobials for
and increases growth of premature children.149 treatment might be on prescription, whereas some
Where authorised, antibiotics used for growth promotion products indicated for prevention are not. In other areas,
can generally be purchased over the counter without regulation and capacity to supervise can be very weak. The
veterinary involvement. In many countries, growth World Organisation for Animal Health (OIE) provides
promoting use of several antimicrobials is authorised and guidance and capacity building, especially in those areas.
widely practised. In the EU, restricted authorisation of Data for amounts of antimicrobials sold for or used in
antimicrobial types began several decades ago. On the animals are still scarce in most regions of the world. The
basis of recommendations by a committee chaired by network for European Surveillance of Veterinary
For more on ESVAC see http:// Professor Michael M Swann, the UK withdrew Antimicrobial Consumption (ESVAC) was formed by the
www.ema.europa.eu/ema/index. authorisation for growth promotion of several substances European Medicines Agency on request by the European
jsp?curl=pages/regulation/
document_listing/document_
including tetracyclines and penicillin in 1971.150 The EU Commission to collect comparable data for consumption
listing_000302.jsp and neighbouring countries followed suit in the 1970s. of antimicrobials for animals in the EU. Only data
Sweden banned the all use of antimicrobials for growth aggregated for all animal species are collected. To correct
promotion in 1986, and Denmark, Finland, and Norway for differences in animal populations over time and
abandoned all such use in the late 1990s.151 Finally, all between countries, a population correction unit was
growth promoting use was abandoned in the EU in 2006. developed, roughly equal to the estimated live weight of
In the USA, the Food and Drug Administration (FDA) has the animal population in each country. Data for 2011
released draft guidelines on judicious use of antimicrobials suggest substantial differences in amounts sold in
in the rearing of animals for food production. These participating countries.156 Many factors might explain this
recommendations aim to reduce the overall use of finding, including a blunt unit of measurement and
medically important antimicrobials and include veterinary differences in composition of the animal populations and
oversight and consultation. If this guidance is adhered to, a systems for production and disease profiles between
gradual phasing out of growth promoting use is to be countries. Furthermore, many of the participating
expected. countries have only recently set up systems for data

1068 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

collection and at least a couple of years are probably needed contaminated poultry meat has been implicated as a
to establish a good baseline. However, other explanations source of E coli causing urinary tract infections in women.163
must exist, such as differences in ways to prevent diseases Less understood is transfer indirectly via the environment.
and prescription behaviour. In most countries, products After land application of manure, salmonella and
are mostly intended for medication of groups of animals campylobacter will survive for some time in soil, depending
via feed or water. Overall, tetracyclines, sulphonamides, on the environmental conditions.164 Contamination of
and penicillins were the main classes sold. In the EU, vegetables and other crops directly from soil or through
monitoring of resistance in commensals from healthy irrigation with contaminated water is a possible but poorly
animals are reported to the European Food Safety documented route of spread.
Authority; when figures on sales of antimicrobials are Direct evidence of spread of resistance genes between
compared with figures on resistance—in E coli to the microbiotas of various animal species and people is
tetracycline, for example—countries reporting lower sales difficult to obtain, and the precise routes of spread more
also report lower prevalences of resistance.156,157 However, difficult to discern. However, indirect evidence clearly
because the use of antimicrobials varies between animal suggests that such transmission does happen. Findings
species and even between production systems, further from an early experimental study68 showed spread of an
associations between sales and resistance are hampered by E coli multiresistance plasmid between chickens and from
the dearth of sales data by animal species. A process chickens to people in contact with the animals. On a
aiming to set up systems for harmonised collection of population level, studies of dissemination of genes
sales data by species and to develop more refined units of conveying resistance to antimicrobials used in animals but
measurement is underway within ESVAC. not in people or vice versa provide strong indications of the
direction of spread. In former East Germany, a
Complex pathways streptothricin antimicrobial was introduced for growth
The interface between human beings and animals is promotion in 1983. Shortly after the introduction, a
complex; numerous possible pathways exist for streptothricin-resistance gene carried on a transposon was
transmission of resistant bacteria. The fact that resistance recorded in E coli from pigs. Subsequently, the resistance
genes can be transferred between different commensal gene appeared in E coli from farmers, their family
bacterial species and from those to pathogens adds to the members, from urinary tract infections of urban citizens,
complexity. Exposure through food is the most commonly and later also in salmonella and shigella isolated from
studied transmission route and the most important. The cases of diarrhoea in people.153 Another example of putative
most likely source of resistant bacteria in food of animal spread is the gene aacC4. This gene conveys resistance to
origin is contamination from animals’ intestines during apramycin, gentamicin, and tobramycin in Entero-
slaughter, but there are numerous other stages in food bacteriaceae. Apramycin is an aminoglycoside used only in
production where contamination with microbes, or animals, mostly mixed in feed. The gene aacC4 gene has
amplification or reduction of their numbers, can happen been identified in E coli and Salmonella enterica serotype
(figure 5). Furthermore, exchange of resistance genes Typhimurium (S typhimurium) from animals and their
between bacteria from different sources can happen at all environment, but also to a limited extent in people.165
stages, including in the kitchen.158 Food is traded The detection of vancomycin-resistant enterococci with
internationally, which means that local production does the vanA-gene cluster in animals in the EU triggered
not equal local consumption. Epidemiological studies of
food-borne transfer of antimicrobial resistance sometimes
generate conflicting results. Less explored are potential
environmental routes.159 Manure and biological solids
applied to land might contain both antimicrobials and
resistant bacteria. Through run-off from fertilised land or
directly from sewage, contamination of surface water can
also occur. Spread to human beings and animals is
possible through contact with soil, irrigation of crops,
contact with water, or with wildlife.

Evidence of spread
Ria Novosti/Science Photo Library

Direct spread of MRSA from animals to people in close

contact is well documented.160 Transfer of community or
hospital-associated MRSA from people to animals has also
been reported.160 Food-borne transmission of non-typhoid
salmonella and of campylobacter from animals is well
established and arises whether bacteria are resistant or
not.161,162 More recently, handling or consumption of Figure 5: High standards of food processing can prevent contamination of food with bacteria

www.thelancet.com/infection Vol 13 December 2013 1069

 The Lancet Infectious Diseases Commission

several investigators to explore possible transfer from The effect

animal to human enterococci. Although some types of A direct effect of emergence and spread of antimicrobial
the gene cluster are found only in people, an overlap has resistance in bacteria of animal origin is the loss of
been reported between types found in isolates from effectiveness of antimicrobials used for treatment of
animals and from human beings.153 In an experimental animals. This aspect is poorly documented, but without
study, Lester and colleagues166 showed transfer of the effective treatment of serious diseases, mortality and
gene cluster from ingested animal to human enterococci morbidity would increase with negative effects on animal
in the intestine of three of six healthy volunteers not welfare. In the rearing of animals for food production,
receiving antimicrobials. Thus, transient carriage of there would also be consequences for productivity and
animal-derived commensals can result in transfer of economy. These effects are not limited to intensively
resistance genes to bacteria likely to be better adapted to reared animals. In LMICs, infectious diseases in animals
human hosts. can have a substantial effect on the economy of a local
The most challenging question hitherto is to what extent community dependent on small-scale rearing of
animals contribute to the spread of genes conveying ESBL animals.155 Eventually, the spread of resistance could lead
or plasmid-mediated AmpC-type enzymes. Because ESBL to a local food security problem with negative effects on
and AmpC production can be conveyed by many different public health.
genes, which can be carried on different plasmids, many More studied, and far more controversial, is the potential
possible permutations exist. Occurrence of E coli producing effect of spread of bacteria and resistance genes from
various ESBLs or AmpC in companion animals, horses, animals on public health. Guidance for risk assessment of
and animals reared for food production is increasingly food-borne antimicrobial resistance has been agreed by
reported.167 In several studies, isolates from animals or the Codex Alimentarius Commission.171 A full quantitative
food products have been compared with isolates from risk assessment in this area requires large amount of data
human beings in the community or in hospitals. In a study including relevant endpoints, such as public health burden
from the Netherlands,168 20% of the isolates from people of resistance. Because of the many possible routes of
carried any of the two most commonly occurring plasmid- transmission and the complexity of resistance epi-
gene combinations occurring on broiler meat on the demiology, different attempts to estimate the risk have
domestic market, suggesting transmission via food. By yielded quite different results. Environmental aspects
contrast, in a study from Sweden,169 the overlap between could be taken into account in risk assessments related to
the plasmid-gene combination that dominates in broilers antimicrobial resistance.152,172 Models made with such an
raised in that country was rare in clinical isolates from approach could provide a more holistic understanding of
human beings. The reasons for the differing results are not the importance of different effects of use of antimicrobials
known, but various context-specific factors along the farm- and in the identification of the most crucial control points.
to-fork chain might be implicated—eg, differences in WHO has developed a list ranking antimicrobial classes
amounts and types of antimicrobials used in broiler according to their importance for public health. Two
production. criteria were used: whether the drug is the sole treatment
An investigation of the temporal patterns of occurrence or one of few alternatives to treat serious human diseases,
of various genes conveying ESBL or AmpC-production or used to treat diseases caused by organisms that might
suggests that in many cases, emergence in human be transmitted via non-human sources or might acquire
medicine seems to pre-date emergence in various resistance genes from non-human sources. Anti-
animals. For example, CTX-M-15 is one of the most microbials that meet both these criteria are classified as
commonly reported ESBLs in human beings worldwide, critically important. The WHO list can be used to support
but is rarely reported in animals reared for food decisions about data collection, risk assessment, and risk
production.167 By contrast, CTX-M-15 is commonly management. To further support allocation of resources,
reported from wild birds.159 At least some of the genes WHO has prioritised critically important antimicrobials:
conveying ESBL or AmpC might originally have been fluoroquinolones, third and fourth generation cephalo-
introduced in various animal populations from human sporins, and macrolides are viewed as the highest priority
carriers, maybe through environmental routes. If so, for risk analysis.173
they have subsequently been spread between animals, Although the exact effect of transfer of resistance genes
farms, and regions leading to the present situation. or bacteria in any given context remains diffuse, food
Recent occasional findings of carbapenemases in borne pathogens and MRSA are spread from animals to
Enterobacteriaceae originating from animals reared for people. Strong circumstantial evidence suggests that
food production170 might also be explained by introduction resistance genes circulate between people, animals, and
from other sources. Carbapenems are not authorised for the environment. Any further increase in prevalence
use in animals, but through the use of any β-lactam or among animals will increase the likelihood of spread to
other antibiotic to which the carrying bacteria is resistant, other realms. Additionally, even rare transfer events could
these genes can now be amplified through co-selection have a substantial effect if secondary amplification takes
and spread in animal populations. place in hospitals and the community.174 Veterinary

1070 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

medicine and agriculture need to apply antimicrobial lead to rolling back major achievements in modern
stewardship to curb further emergence and spread of medicine.176 Because resistance inevitably follows antibiotic
antimicrobial resistance in the sector. use, the paradox is that populations—both in industrialised
countries and LMICs—can face challenges of access and
Time to move on from blame and shame excess. Even in high-income countries, antibiotic use
The use of antimicrobials for animals reared for food ranges widely, with three times greater outpatient
production remains controversial. Knowledge gaps in the consumption in Cyprus than in the Netherlands.177 The
understanding of the broader resistance dynamics, evidence points to increasing use of antibiotics in hospitals
conflicting results in various studies, and in particular the over time, much of it not consistent with clinical
difficulty quantifying the potential effect on public health guidelines.122 Although some patients are prescribed
are important barriers for change. This situation has led to unnecessary courses of antibiotics, others are not given
a polarised debate, in which some state that the effect is appropriate treatment.
small, while others argue that there is major harm. This challenge is one of access and rational use, as
Policies vary between countries and regions, and are not defined as “how to ensure that when [patients] need drug
restricted to antimicrobials for animals, animal health and therapy the appropriate drug is prescribed for them, it is
welfare, and food safety. Food is traded globally and other effective and of acceptable quality and safety, it is available
factors such as economic, export, and trade policies can at the right time at a price they can afford, it is dispensed
play a part, as can consumers’ expectations of access to correctly and it is taken in the right dose at the right
affordable quality food products. Interests in these areas intervals and for the right length of time”.178 With this
might well be conflicting, and this is important to definition in mind, challenges associated with access of
acknowledge in the debate and in decision making. antibiotics begin with a recognition of the right to health,
General guidance spanning regulatory needs and but result from therapeutic, financial, and structural
prudent use of antimicrobials is provided by all international barriers. Therapeutic access refers to the bottlenecks—
organisations: the OIE, WHO, and the UN Food and scientific and financial—in bringing new antibiotics to
Agriculture Organization (FAO). Implementation of market. Financial access characterises the difficulty in
prudent use relies on the daily work of farmers and affording a rational course of antibiotic treatment, and
veterinarians. Legitimate conflicting interests can surround structural access addresses the obstacles to delivering
this implementation—eg, production economy and the antibiotics effectively in the system and using them
ethical obligation to care for diseased animals. In view of rationally at the clinical level in-country. Each of these
the polarised debate, veterinarians and farmers might feel barriers to access need to be surmounted if an antibiotic is
that they are blamed for a problem they perceive is to progress from bench to bedside. These barriers are also
essentially generated by medical doctors.175 This situation shown in the so-called glocalisation of antibiotics (ie, how a
might lead to a defensive attitude and does not cater for global product integrates into local markets). Delays in the
productive solutions. A way forward would be to entry of novel antibiotics—and delays in the availability of
acknowledge that human health, animal health, and the complementary technologies like diagnostics or vaccines—
environment are all interlinked, and that the responsibility can affect the access and use of antibiotics in a local market.
for dealing with the problems of resistance is shared by all Importantly, excess or overuse also contributes to
stakeholders. Strong local and global partnerships are problems of antibiotic access. Antibiotics are overused for
needed in which policy makers, academia, and many reasons, including patients’ expectations of pre-
professionals from all sectors work together to improve scribed treatment; information asymmetry at the user,
present systems. The common goal should be to preserve prescriber, or provider levels; diagnostic uncertainty; and
the effect of antimicrobials for future generations of human the many financial incentives for overprescription, all of
beings, but also for animals. Antimicrobials should only be which drive both presumptive and unnecessary use of
used when needed. In the case of animals, this means that antibiotics. The excessive use of antibiotics traces, in part,
growth promotion and routine prevention with anti- from the revenues their use generates for health systems
microbials also used for treatment should be phased out, as and health-care providers. In China, the dispensation of
recommended by the Swann Committee.150 When needed, drugs comprises a substantial part of provider incomes,
antimicrobials should be used judiciously. Furthermore, and data from 28 cities reveal a prescription rate of
long-term efforts are needed with a focus on reducing the antibiotics twice that recognised as appropriate by WHO.179
need for treatments by improving infection control and Such financial incentives exist at all levels of the system,
management and by developing robust systems. from pharmaceutical companies to procurement agencies,
retailers, prescribers, and dispensers, and include the
Part 5: The access and excess dilemma informal market.
A global balance
To tackle antibiotic resistance needs not only a renewal of Strategic points for intervention
the depleted pipeline of novel antibacterial drugs, but From bench to bedside, strategic points for intervention
conservation of those now in use. A failure to do so might exist to address access and excess in antibiotic use. The

www.thelancet.com/infection Vol 13 December 2013 1071

 The Lancet Infectious Diseases Commission

interaction among prescribers, dispensers, and patients is expenditure on diagnostics in LMICs is similarly low.188
central in this value chain. However, surrounding this Product development partnerships like the Foundation for
For more on FIND see http:// chain is a health-care delivery system with many points at Innovative New Diagnostics (FIND) and Partners
www.finddiagnostics.org/ which policy leverage might positively or negatively affect Advancing Transitions in Healthcare (PATH) have made
For more on PATH see http:// access to antibiotics.180 important contributions to bringing diagnostics suited to
www.path.org/ To balance access and excess requires reconceptualisation low-resource settings to market. Notably, FIND supported
of the delivery of antibiotics as a complex adaptive system. the development of the GeneXpert MTB/RIF system,
The interaction between the components of such a which not only identifies patients with smear-positive
complex system is non-linear. Product, information, and tuberculosis, but also establishes resistance to rifampicin
financing flow through the value chain of pharmaceutical from untreated sputum samples in less than 2 h.189 Product
delivery of antibiotics. Adequate regulation of these flows profiles for point-of-care testing that returns susceptibility
through the value chain can balance access and excess use results in hours, not days, can provide a road map for
of antibiotics in the delivery system. The value chain targeted grand challenges in diagnostics research for
includes multifaceted innovation, dissemination and bacterial pathogens.
introduction of new techniques, technology, and rules, The assessment of syndromic management and the
scaling up and implementation of these approaches, and clinical algorithms based on existing diagnostic methods
then assessment and monitoring of their effect on access have to complement the development of new diagnostics.
and excess of antibiotics. A clinical algorithm now allows community health workers
to make a presumptive diagnosis of acute lower respiratory
Innovation in tackling antibiotic resistance infection, but had a better diagnostic been available,
With the dearth of novel antibiotics, scientific bottlenecks unnecessary treatments might have been avoided.190 Even
in the research and development pipeline need to be where access to such technologies exists, distrust in the
overcome, from identification of promising leads and quality of diagnostics, the paucity of timely results from
crossing from basic and translational research to diagnostic tests, and the fear of poor outcomes, can prompt
improvement of clinical trials without sacrificing safety and clinicians to set aside diagnostic test findings. Improved
targeting reimbursement.181 To ensure access, but not diagnosis—part technology, part syndromic manage-
excess, investment in research and development should be ment—can reduce uncertainty about whether to treat with
unlinked from returns. Measures such as extended data antibiotics or not. Taken to scale, the advent of simple
exclusivity and premium pricing are not the answer. Such diagnostic tests for yaws and trachoma could help reduce
approaches tie revenues to volume-based sales, and worse the need for mass administration of antibiotics, such as
yet, higher prices place access to those in need at risk. azithromycin.191–193 Improved diagnostics can help convert a
Alternatively, public funding could be conditioned to buy vicious cycle into a more virtuous one.
out patents. Manufacturers could then be licensed to
produce antibiotics on a scale appropriate for rational use.182 Dissemination and introduction of antibiotic treatment
Innovation is needed, not only for the development of Antibiotics, costing as little as $0·13 to $2·03 per patient,
new antibiotics, but also for combination therapy. Of could save an estimated 509 000 lives from newborn sepsis
course, there is the synergistic action of using drugs in and pneumonia every year.194 In fact, fewer than a third of
combination, such as trimethoprim and sulfamethoxazole children with suspected pneumonia in LMICs receive
in co-trimoxazole, amoxicillin, and clavulanic acid, or even potentially life-saving antibiotics (such as amoxicillin
perhaps someday silver with some antibiotics.183 However, provided twice daily for 3–5 days) for treatment. The gap
there is also the reality that if the probabilities of resistance between children younger than 5 years who receive
to drugs A and B are independent, then the chances of treatment and those who do not can be quite pronounced
developing resistance to both drugs used together will be when a comparison is made between the top and bottom
much less likely—the product of those probabilities.184 By quintile of households on the basis of wealth.195 By contrast,
targeting many mechanisms of resistance simultaneously, antibiotics are too often prescribed for diarrhoea when oral
combination therapy might help slow the emergence of rehydration salts and zinc would better serve patients.196
resistance.185 Access to such treatment contributes to growing
Innovation is also needed for technologies com- community-wide drug resistance. In higher income
plementary to novel antibiotics. Preventive vaccines can countries like the USA, overuse of antibiotics for upper
reduce the need for antibiotic treatment,186 but better respiratory infections not only persists,197 but is also
diagnostics can both accelerate the recruitment of patients implicated in the widespread use of broad-spectrum
with multidrug-resistant infections into clinical trials and antibiotics, for patients treated in ambulatory care.198 The
narrow the use of new antibiotics once on the market. As a increased risk of drug-resistant infection further decreases
method to hold resistance in check, the potential of the clinician’s threshold to prescribe antibiotics more often
diagnostics might be underused. Diagnostics in the USA and with broader spectrum, thereby feeding a vicious
guide 60–70% of health decisions, but might account for cycle. In some places and over the internet, antibiotics can
only 2% of health expenditures.187 The health-care be obtained without prescription.199 Worse yet, substandard

1072 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

antibiotics may be ineffective but still contribute to drug Incentives do not consistently align in a way that promotes
resistance. For those living on less than $1 per day, upfront antibiotic stewardship or rational use. Even low levels of
purchase of a complete course of antibiotics might be resistance can amplify into greater societal costs as
beyond financial reach, and adherence to a full course of clinicians minimise risks to patients by prescribing
prescribed antibiotics is unlikely. broader spectrum, presumptive treatment. Mothers who
Access to antibiotics is not just a matter of innovation in bring children to health centres might seek antibiotics not
product development; even for existing drugs delivery may just for that episode of illness, but stockpile the extra
be wanting, especially in resource-limited settings. For antibiotic for that future occasion when reaching the clinic
example, data from a study in Tanzania200 suggested that might not be so easy—a survival strategy where access is
Gram-negative sepsis in children was associated with a limited. Realignment of incentives to providers,
mortality rate double that of malarial infection. Nearly half prescribers, dispensers, and users is important to
the neonates with Gram-negative sepsis at a tertiary encourage correct use of antimicrobial treatment,
hospital in the Mwanza region had infections resistant to including proper diagnosis, correct choice of treatment,
third-generation cephalosporins. However, almost all of generic prescription, and generic substitution.
the Gram-negative enteric bacteria were sensitive to Another strategic point of intervention might be in
meropenem, a drug too expensive and often unavailable in procurement of antibiotics, in which innovative approaches
many LMICs.201 can provide access and prevent excess. For example, the
Narrowly framed interventions aimed at dissemination facilitation of the supply of second-line treatment for
and introduction of antibiotics and their rational use might multidrug-resistant tuberculosis. Such treatment can be
focus exclusively on health-care providers in hospitals and 50–200 times more expensive than first-line treatment.
clinics (figure 6). This narrow view has focused past efforts The Green Light Committee reviews applications and
on prescription audits and provider training. Stewardship provides technical assistance to countries seeking drugs to
implies both effective treatment of patients with antibiotics treat multidrug-resistant tuberculosis. Companies provide
and minimisation of collateral damage from the use of these drugs at concessionary prices through the Green
these drugs. A meta-analysis of studies122 of antibiotic Light Committee process, with the assurance that the
stewardship programmes suggested approaches that drugs will be used appropriately and in a way that
restrict prescription of antibiotics had larger effect than did minimises resistance.203
persuasive approaches, but that over time, the differences
were not statistically significant. Even if some multifactorial Scale-up and implementation
interventions under study seem more promising than Breakthrough approaches to solve the access and excess
others, one approach is unlikely to suit all settings. Meta- dilemma will need the integration of antibiotic stewardship
analyses can incorporate publication bias that comes from programmes into health-care delivery systems. Without
not reporting negative findings and thus fail to capture the such vision, stewardship programmes might remain
underlying heterogeneity of study settings.94 experimental or pilot interventions without substantial
Effective antibiotic stewardship efforts, however, need effect on antimicrobial resistance or access to life-saving
broader control over product, information, and financing. treatment.
The flow of antibiotic product is in the system might be An effective new diagnostic test for bacterial acute lower
controlled by restriction of formularies, requirement of respiratory infections could save at least 405 000 children’s
preauthorisation of antibiotic use, and de-escalation of
broad-spectrum coverage when a pathogen is identified.
Prospective audits with intervention and feedback,
education of patients, guidelines for providers, and
computer-assisted strategies affect the flow of information
in the system. Approaches such as offering a safety-net
antibiotic prescription for otitis media and instructing
parents not to fill the prescription unless symptoms
worsen or do not improve after 48 h, might help set the
default option in favour of rational use.201 Such an approach
involves re-engineering the process by which health care is
Mauro Fermariello/Science Photo Library

delivered. Additionally, checklists—perhaps through

antibiotic order forms, computer order entry, or infectious
disease consultant approval—can ensure more rational
use of antibiotics.202 The appropriateness of these
approaches is context-dependent and might differ between
LMICs and higher income settings.
From bench to bedside, there are both financial and non-
financial incentives and affects on decision-making. Figure 6: Dispensary staff are key in providing access to antibiotics, but also have a role in preventing excess

www.thelancet.com/infection Vol 13 December 2013 1073

 The Lancet Infectious Diseases Commission

lives per year.190 For example, the value of a rapid diagnostic ivermectin treatment for onchocerciasis has shown,
test for malaria in the setting of fever will vary by local removal of the barrier of drug costs alone is not sufficient
prevalence of the disease, by seasonal changes, and by how to ensure delivery.210 In scaling-up this programme to reach
often another cause of fever, like pneumonia, might 25 million people every year in Africa, Merck acknowledged
coincide with a positive malaria diagnostic test finding. that this donation needed complementary investment to
Local context matters. However, when such testing train health-care workers to integrate such efforts into the
protocols are applied, do health-care providers adjust to the delivery system.211 Most notably, the campaign to eradicate
context? Evidence from the scientific record on the effect trachoma has come to rely on one dose of azithromycin,
of diagnostic testing on prescription behaviour is although this is part of a larger SAFE strategy, including
mixed.204,205 The increased prescription of antibiotics in the surgery for trichiasis and advanced disease, facial
setting of a negative rapid diagnostic test finding for cleanliness to reduce transmission, and environmental
malaria might lead to improved rational use, or it could improvements. In addition to ready-to-use therapeutic
just result in overprescription of antibiotics as food, findings from a recent study212 support the routine
compensatory behaviour to mollify patients’ expectations use of antibiotics as a part of the management of severe
of treatment.206 acute malnutrition, which claims 1 million children’s lives
In the health-care delivery system, antibiotic stewardship every year.
might be through integrated community case manage- A policy tension might arise between saving lives with
ment. Applied to malaria, diarrhoea, and pneumonia, this short-term mass campaigns that advocate antibiotics, and
strategy includes community health workers in the increased mortality as a result of increased antibiotic
management of uncomplicated childhood illness and the resistance. However, investigators piloting such
referral of more complicated cases. Factors, ranging from interventions are already making such calculations. A trial
how compatible programmes are with local sociocultural investigating the effect of mass administration of
beliefs to how secure supply inventories of drugs and azithromycin on the reduction of childhood mortality in
diagnostics are, can affect the introduction of integrated Tanzania, Niger, and Malawi is underway. The study’s lead
community case management.207 In a study in Burkina investigator, Thomas Lietman acknowledged this trade-off:
Faso, Ghana, and Uganda,208 community health workers “We think we will select for antibiotic resistance. However,
were taught to use rapid diagnostic tests for malaria and to many of us believe that might be a price worth paying if
count the respiratory rate to diagnose pneumonia. there is truly a reduction in childhood mortality.”213 The
Although compliance with the test results was high and investigators will carefully monitor the development of
unnecessary use of artemisinin combination treatments antibiotic-resistant diseases in the target communities.
curbed, varying degrees of both antibiotic overuse and Lietman notes that in communities treated for trachoma,
underuse were noted. After a simple clinical algorithm, azithromycin resistance dissipated within 2 years of mass
community health workers in rural Zambia used rapid administration ending. Moreover, azithromycin is seldom
diagnostic tests with few adverse events, reliably dispensed used in sub-Saharan Africa, and resistant infections
artemether–lumefantrine for malaria and amoxicillin for remain susceptible to many classes of other antibiotics.
pneumonia, and effectively managed their supply of The unanswered question is whether this picture of
medicines.209 The dispensing of antipyretics to children infection will remain true in Africa?
with negative results of rapid diagnostic tests also seemed The lessons learnt from a mass campaign differ from
to help meet caregiver expectations. those gained from the treatment of episodic illness.
Bridging the divide between individual and collective However, the episodic treatment of bacterial infections also
action will also be key. The misuse of antibiotics confers needs infrastructure for effective delivery. Treatment of
risks upon the individual, not just the wider community. acute rheumatic fever to prevent carditis and the sequelae
These risks include the selection of resistant and more of valvular lesions includes secondary prophylaxis with
virulent infections, opening the door to opportunistic antibiotics. Adherence to the regimen of monthly,
fungal infections, and leading to subsequent, more drug- intramuscular injections might best be delivered with a
resistant rounds of bacterial infections. Similar trends can control programme, equipped with a central register, a
be seen at the organisational level. Is there a failure of supply of benzathine penicillin G, and appropriate
collective action when no individual insurer will invest in outreach. Additionally, the plans underway for eradication
infection control at a local hospital, in case other insurers of yaws might include a diagnostic test for Treponema
free-ride off that investment? Nor will any individual drug pallidum and allow for targeting an oral dose of azithromycin
company show self-restraint in marketing an antibiotic to those individuals infected.214 The challenge in all these
when its therapeutic competitors do not. Collective action efforts will be to scale-up antibiotic use but to minimise
will need new forms of partnerships, such as pooling of drug resistance from unnecessary or inappropriate use.
the insured risks for infection control at a hospital.
Antibiotics are not just part of routine episodic care in Monitoring and assessment of access and excess
the prophylaxis and treatment of bacterial infections. As Successful antibiotic stewardship programmes need
the experience of population-wide campaigns such as effective feedback loops to be established in health-care

1074 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

systems and mobilisation of data as triggers for follow-on challenge competition could also be the effective
action. Taking stock of these programmes, quality development and deployment of technologies com-
measures, such as time to first antibiotic for a patient with plementary to antibiotics, notably diagnostics. To
community-acquired pneumonia, are useful. Some have disseminate and introduce these technologies, antibiotic
argued such programmes should be assessed on outcomes, stewardship might have to go beyond audit and feedback
not just process measures. Although relating specific to realignment of incentives. This might also require
measures causally to antibiotic stewardship programmes strategic points for intervention beyond providers,
can be challenging, infection-related mortality, hospital prescribers, and patients, such as at the level of
length of stay, readmission rates, C difficile infection rates, procurement of antibiotics, to be found.
and antibiotic resistance levels can be useful outcome The scale-up and implementation of these interventions
dimensions to track.215 Just as vital signs like blood needs infrastructure, not unlike that of mass administration
pressure, heart and respiratory rate, and body temperature campaigns, but as exemplified in efforts to treat acute
give a snapshot of a patient’s illness, select quality rheumatic fever, these might need longer term
measures can give a similar picture of how effectively an commitments to tracking and follow-up of patients such as
intervention is tackling antibiotic resistance in the health- in a central registry. Scaling up is also associated with
care delivery system. Data from such surveillance can rethinking how incentives work when moving from
provide useful feedback to antibiotic stewardship individual to collective action. New institutional
programmes, which can be redesigned to take stock of arrangements might be needed to pool and share the costs
positive, negative, intended, and unintended effects and benefits of these interventions across providers,
highlighted in thorough monitoring and assessment. insurers, or drug companies. Finally, monitoring and
Trends of antibiotic use can be mapped,216 and perhaps assessment requires effective feedback loops that
the discovery of regional or local differences might transform surveillance into follow-on action. Learning
encourage responsible parties to act. However, linking collaborations of like-minded individuals or institutions
surveillance to action can be challenging. The Institute for are one promising approach to motivate change and to
Healthcare Improvement applies a continuous quality share lessons from such efforts.
improvement approach, consisting of plan–do–study–act Access to new antibiotics will need to be controlled
in its work.217 To tackle antibiotic resistance, collaborations through strong regulatory, procurement, and distribution
involving like-minded groups can encourage the sharing oversight while existing products will need to be protected
of such continuous quality improvement lessons among and their shelf-life extended through coherent and
local innovators. A process of collaborative improvement inclusive approaches to antibiotic resistance. Suggested
among institutions facing similar difficulties was the solutions to this major issue lie in interventions aimed at
hallmark of the Institute for Healthcare Improvement’s re-engineering the delivery system. Public health services
breakthrough collaborative.218 As part of its state-wide can partner with and actively promote accredited drug-
Antibiotic Stewardship Program Initiative, the California dispensing outlets, where rational use might be strictly For more on the Antibiotic
Department of Public Health has also set about developing audited; rational use might be enabled through vouchers Stewardship Program Initiative
see http://www.cdph.ca.gov/
collaborations among hospitals that share common delivered through mobile-phone-based money transfers. programs/hai/Pages/Anti
challenges.219 Various approaches to ensure that the full course of microbialStewardship
antibiotics is taken, not just dispensed, need to be piloted. ProgramInitiative.aspx
Achieving access without excess These scenarios are only examples of system-wide
For the 1 billion people who live on less than $1·25 a day, a approaches that aim at finding the right balance between
full course of antibiotics paid for upfront might not be access and excess.
affordable. The balance of access and excess poses an
important ethical question as to who should access Part 6: Challenges of antibiotic resistance in weak
antibiotics, under what circumstances, and who should health systems
make these decisions if they should not be left to the An underestimated burden
discretion of users, or even prescribers and providers. If Over the past decade, antibiotic resistance has risen
antibiotics are treated as a non-renewable resource, access alarmingly worldwide.220 Among the key players, including
to both new and existing products needs to be thought global health donors, pharmaceutical companies, technical
about carefully. agencies, and governments,221 patients and physicians have
Taking a systems approach, tackling antibiotic resistance the strongest effect on resistance rates, because selection
involves intervention at the level of innovation, dis- and spread of resistant organisms is mainly a local process
semination and introduction, scale-up and imple- based on practices in individual hospitals and
mentation, and monitoring and assessment. Every stage communities.222
offers the opportunity to re-engineer the system. To In LMICs with weak health systems, the effect of
develop new antibiotics, financing models that delink antimicrobial resistance on health and economics is
research and development investment from revenue largely underestimated and incompletely understood. At
returns are needed. The productive focus for a grand least two-thirds of childhood mortality is related to

www.thelancet.com/infection Vol 13 December 2013 1075

 The Lancet Infectious Diseases Commission

infections, and children are therefore probably more Diarrhoeal pathogens

vulnerable than are adolescents and adults.223 The increase Vibrio cholerae, salmonella, shigella, E coli, and
in resistant strains among bacteria causing some of the campylobacter accounted for 500 000 diarrhoeal deaths in
commonest childhood infections (eg, newborn sepsis, 2010.246 Cholera is still a public health priority in most
meningitis, pneumonia, diarrhoea, and typhoid) can make LMICs; however, resistance to the recommended
clinical outcomes worse and render first-line empirical antibiotics (co-trimoxazole, chloramphenicol, sulpho-
antibiotic regimens ineffective. namides and nalidixic acid) has been reported from
Therefore, the challenge lies not only in the use of countries in South Asia and Africa.247–251 In most of sub-
existing knowledge to make antimicrobial resistance a Saharan Africa and Bangladesh, resistance to co-
national health priority, but also in the implementation of trimoxazole and furazolidone is increasing, whereas
customised containment strategies.90 These strategies resistance to tetracycline fluctuates from year to year.252–254
must, however, successfully integrate scale-up of proven Resistance and reduced susceptibility against tetracycline,
interventions at the service delivery level with a sustainable ciprofloxacin, and azithromycin have also been
self-evaluation process whereby action plans are reviewed reported,248,255,256 suggesting the importance of epidemiology
and revised on the basis of collected evidence.224 to guide treatment.
In LMICs, Shigella flexneri is the commonest cause of
Antibiotic resistance in priority pathogens bacillary dysentery in children younger than 5 years.
Neonatal sepsis Resistance to ampicillin, tetracycline, co-trimoxazole, and
Studies of newborn sepsis from Bangladesh, Ethiopia, chloramphenicol is common in developing countries.257 In
Nepal, Tanzania, Cameroon, Ghana, India, Nigeria, Africa and Asia, resistance against ciprofloxacin—the
Pakistan, Yemen, Vietnam, Philippines, and Egypt,40,225–235 WHO recommended antibiotic for bacillary dysentery in
support trends in pathogens and resistance highlighted by children—rose from 0·6% to 30% over 10 years.258 Further,
Waters and co-workers14 and Zaidi and colleagues.236 The reports of emerging resistance to second-line antibiotics
high proportion of resistance among Enterobacteriaciae such as third-generation cephalosporins and azithromycin
(Klebsiella spp and E coli), the leading cause of newborn are of great concern.259
sepsis in developing countries, to ampicillin, gentamicin, Increased resistance to flouroquinolones and macrolides
ceftriaxone, and ciprofloxacin, is concerning. Neonatal is also being reported for campylobacter. Ciprofloxacin
nurseries in some LMICs report meticillin resistance rates resistance rates range from 65% to 88% in Bangladesh,260
of 17–28% in S aureus isolates.40,216,237 Similarly, Lubell and whereas rates of resistance from African countries range
colleagues238 recorded poor susceptibility to almost all from 5% to 38%.257,261 Antibiotic use in veterinary medicine
commonly used antibiotics in pathogens such as S aureus can cause fluoroquinolone resistance in campylobacter.262
and Klebsiella spp. Without rigorous evidence, most Multidrug resistance (to ampicillin, chloramphenicol, and
clinicians agree that 90% susceptibility or higher of co-trimoxazole) in S enterica serotype Typhi (S typhi) is
potential pathogens to antibiotics in critical care settings associated with increased severity and high case fatality
like neonatal intensive care units, and 80% or higher in rates of typhoid fever.261 Fluoroquinolones, macrolides, and
non-critical health-care settings, is acceptable. cephalosporins are the second-line, albeit costlier,
regimens for treating typhoid fever. However,
Pneumonia and meningitis fluoroquinolone resistance is increasing in India (44%)
Streptococcus pneumoniae, a leading cause of community- and Pakistan (58%), and susceptibility is decreasing in
acquired pneumonia and bacterial meningitis in children Congo (15%) and Cambodia (80%).263–266 In Kenya,267 more
is increasingly resistant to macrolides, third generation than 77% of the S typhi were multidrug resistant, much
cephalosporins, and fluoroquinolones. Although higher than the reported 52% in Ghana268 and 29% in
penicillin non-susceptibility of S pneumoniae (PNSP) Egypt.263 Additionally, the proportion of S typhi in Kenya
does not seem to affect outcome of antibiotic treatment that are multidrug resistant and resistant to nalidixic acid
in patients with pneumococcal pneumonia,239 PNSP with decreased susceptibility to fluoroquinolones had
might be of great clinical significance in deciding risen from 1% in 2000 to nearly 25% in 2008.
treatment options for life-threatening infections such as Surveillance for resistance trends, monitoring rational
meningitis. Prevalence of PNSP ranges from 0% in use of antibiotics, and development of clinical manage-
Pakistan240 to 44% in Algeria,241 amoxicillin resistance ment guidelines to standardise treatment can curtail
from 20% in Algeria, Morocco, and Tunisia,242 to 46% in development of resistance against the remaining mainstay
Malta,241 and cefotaxime resistance from 8% to 17% in of treatment, third-generation cephalosporins.
North Africa.242,243 Macrolide resistance is as common as In sub-Saharan Africa, non-typhoidal salmonella are the
33% of isolates in Venezuela and 38% in Mexico.244 dominant contributors to invasive bacterial disease. These
Fluoroquinolone resistance is found in 3% of isolates in bacteria are the second commonest cause of neonatal
Pakistan240 and up to 24% in Bangladesh.245 Dual non- meningitis and the third commonest cause of bacterial
susceptibility to both amoxicillin and erythromycin was meningitis in children older than 2 months in Malawi and
reported in 30% of isolates in one North African cohort.242 Kenya.269,270 However, data from some reports suggest

1076 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

S typhi is the main serotype in Africa.267,271 Multidrug by regulatory agencies are needed to control the sale and
resistance and decreased susceptibility to fluoroquinolones supply of expired and counterfeit drugs.
are now widespread among non-typhoidal salmonella,
causing invasive bacterial disease in Kenya and Malawi,272–274 Lack of antibiotic stewardship and poor infection control
and other parts of sub-Saharan Africa.275–277 This situation Absence of antibiotic stewardship is associated not only
poses a major challenge to treatment and management of with emergence of resistance, but also with poor outcomes
disease. In a study from the Democratic Republic of the for patients.281,282 Apart from appropriate laboratory
Congo,278 rates of multidrug resistance varied from 37% to infrastructure, both stewardship and infection control
80% between different provinces. 10% of the S typhimurium require teams of a clinical infectious disease physician,
strains isolated from Kinshasa were also resistant to clinical pharmacist, and committed hospital admini-
azithromycin. Recently, a very high rate (48%) of stration, which are scarce in LMICs.
cephalosporin resistance has been reported from India, Furthermore, insufficient infection control surveillance
especially in S enterica serotype Agona.279 systems within hospitals in LMICs12 with inherent
problems (eg, overcrowding and insufficient equipment
Risk factors for resistance in childhood disease pathogens and trained personnel) leads to spread of nosocomial
Antibiotic misuse infections and even outbreaks caused by resistant
User-related factors such as self-medication, non- pathogens.13 This scenario has huge financial implications
compliance, misinformation, and advertising pressures in for developing countries and can lead to exponential
combination with other factors such as ignorance, lack of increase in treatment costs besides increase in morbidity
education, and inaccessibility to health care and diagnostic and mortality.12,283,284 These resistant pathogens become a
facilities, are the major drivers of resistance. Poverty commensal reservoir of resistant genes and are spread to
compounds the problems, because patients do not have the community through unsafe water and poor sanitation.
access to clean water and hygiene and are at an increased A history of hospitalisation is an important risk factor for
risk of acquiring infections. Individuals living in poverty acquisition of resistant infection in family members.284
also have poor baseline nutritional status to fight off Infection control and antibiotic use are inter-related and
infections and are at risk of treatment termination due to the individual contribution by each of these factors is
affordability issues.280 Other factors relating to health-care difficult to seperate.285
providers that may lead to inappropriate prescribing
include insufficient training, unprofessional conduct, and Paucity of surveillance information
paucity of diagnostic facilities, leading to incorrect Surveillance data are essential for providing information
selection of antibiotics.280 on trends and magnitude of resistance. Unfortunately,
Inappropriate prescriptions resulting from economic timely availability of such data from developing countries
incentives offered by pharmaceutical companies and is scarce.286,287 Causes of this data lag and lack include
inaffordability of appropriate dose and duration of technical constraints such as non-existence of data
antibiotic regimens perpetuate a vicious circle of collection and analysis infrastructure, poor laboratory
suboptimum treatment leading to antimicrobial resistance. infrastructure, and weak leadership and governance to
Another cause of antibiotic misuse among prescribers and recognise resistance and its public health (clinical,
patients is lack of awareness of the health priority status of financial, and pharmacological) implications. Absence of
antimicrobial resistance among every tier of health-care essential epidemiological data leads to delayed or
personnel in developing countries.280 suboptimum revisions in guidelines, and strengthens the
The highest burden of deaths caused by infectious vicious circle of injudicious use of antibiotics by prescribers
diseases is in south Asia and sub-Saharan Africa.232 on the basis of anecdotal or non-evidence-based
Countries in these regions also report a high prevalence of experience.287 Local antibiograms with pathogen-specific
multidrug-resistant pathogens with poverty of resources susceptibility data are crucial in choosing the best empirical
contributing to poor containment of resistant organisms antibiotic in resource-constrained settings287 and are
in hospital and community settings,13 and inadequate heavily rely on quality-assured laboratory support and an
training of prescribers and laboratory personnel,280 intact information system.
contributing to rising resistance through inappropriate
empirical antibiotic choices, and laboratory diagnostics.224 Paucity of leadership and governance
Without appropriate oversight, policy makers are unable to
Quality of antibiotics assimilate known information about global and regional
Poor countries suffer from lack of regulations for drivers of resistance and its public health implications or
pharmaceutical products, leading to availability of to recognise the problem as a national health priority. As a
counterfeit and low-quality antibiotics. Use of these result, policy makers will struggle to propose and
ineffective antibiotics amplifies resistance and leads to implement sustainable, multidisciplinary, and multitier
purchase of more potent and expensive antimicrobials to (pharmaceuticals, food and agriculture, human resources,
treat resistant strains.100 Policies for appropriate supervision financing, and information systems) strategies that link

www.thelancet.com/infection Vol 13 December 2013 1077

 The Lancet Infectious Diseases Commission

science to practicality.288–290 Corruption within health cash flow into resources like information systems,
systems is of particular importance in LMICs where it can laboratory infrastructure, and personnel training.
lead to siphoning off of smaller and therefore more Systems-support guidelines provide information on
precious health budgets. Poor record keeping and general appropriate indicators (ie, scientific, population, and
human-resource mismanagement further add to systems data) for development of antibiotic resistance
inaccurate burden estimation. Furthermore, because of information databases.302 The central taskforce must
the lack of transparency and accountability, countries with standardise laboratory reporting of resistant pathogens in
the most need might miss out on opportunities to attract an electronic programme through liaison between local
donors and strengthen their health systems.291 microbiologists, infectious disease specialists or general
clinicians, and information technology experts. Fraser and
Combating antimicrobial resistance colleagues303 reported successful implementation of
An effective strategy to combat resistance needs action and electronic medical systems as pilot projects in, Haiti,
involvement of individual institutions that commit to Malawi, and Peru. This capacity building can be taken up
implementation of antibiotic stewardship programmes, as a systems-strengthening project with partial com-
invest in their development and operation, train existing or mitment from both individual governments and com-
employ essential staff, and maintain relevant records.220,221 petitive bidding for external funding. The task force needs
In countries with weak health systems, hurdles such as to provide software programs to hospitals across the
insufficient funds, inadequate infrastructure and country so that nosocomial and community pathogens can
management, shortage of trained personnel, non-existent be identified and resistant clones tracked at district,
or poorly implemented infection control policies, and poor provincial, and national levels.87
surveillance records292 need to be addressed and overcome. Taskforce finances need to match annual agendas. Global
Although control and containment strategies should experts, local legislators, financial experts, epidemiologists,
mainly target individual hospitals and communities, public health experts, and private–public sector
vision and direction should come from the platform of representatives can be stakeholders in the finance and
national health ministries. In 2012, medical societies in ensure the best use of resources for evidence-driven
India put together an implementable road map for strategies. Optimum coverage of BCG, diphtheria, pertussis,
resistance containment in the country that pushed for Haemophilus influenzae b, pneumococcal and measles
central coordination and implementation support from vaccines, now available in most developing countries with
For more on GAVI see http:// the government.293 Therefore, physicians can play an Global Alliance for Vaccines and Immunisation (GAVI)
www.gavialliance.org/ important part in advocacy for containment of resistance support, can contribute to reduction in antibiotic resistance
and give relevant technical input to a national taskforce. by reducing disease occurrence and thus the use of millions
In addition to setting national objectives and goals, of doses of antibiotics.304,305 Public–private partnerships have
functions of the taskforce can include seeking and been suggested as a possible way for non-GAVI eligible
maintaining global partnerships with potential funding countries to bridge the vaccination gap.306 Imperfect
agencies, engaging relevant ministries (agriculture, vaccines targeting drug-resistant pathogens have also been
pharmaceutical industry, information, and education) for suggested as a method to improve the population-wide pro-
support functions, regular customisation of the national portion of drug-resistant versus drug-sensitive strains.307
antibiotic policy by review and feedback of surveillance Overall use of antibiotics can decrease with diagnostic
data, and pushing for legislation through advocacy.87 information.52 Validation of point-of-care tests for
Taskforces can customise available guidelines prepared respiratory tract infections, the leading cause of
by local and global experts to fit individual country overprescription of antibiotics in children, might be an
contexts.220,294,295 important step towards improved antibiotic use.135
Some success stories from LMICs are encouraging. For Education and training incentives to health-care
example, Burkina Faso and Ghana have created confidence providers such as prescribers and dispensers in the
in global donors, inviting external investment in community and members of infection control and
strengthening systems through committed steps that therapeutic committees within hospitals have been
ensure transparency and accountability in management of recommended.224 Direct education of patients also
public finances.296 Honduras and Chad have used public improved antibiotic compliance and decreased use of
expenditure tracking surveys to understand how resource unindicated drugs in Peru.224 Training at all levels of a
flows in health systems can be abused.297,298 In Mexico, a multilevel health system should be in conjunction with
transparency and access to public information law was systems-strengthening steps such as provision of necessary
successfully implemented in 2002,299 which paved the way resources (eg, printed guidelines, data systems, and
for a citizens’ audit into funds redirected from an HIV reporting protocols).
programme.300 For other countries such as Pakistan, with Important steps to contain resistance at the primary
the most deterioration reported between 1996–2004,301 health-care level include country-wide scale-up of
deeper systematic reform could be the answer.295 integrated management of treatment protocols for
Championship at the central level is essential to improve neonates and children,308 optimum vaccination coverage,305

1078 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

development and dissemination of education material for range of available classes. Ideally, new classes should be
practitioners and patients on rational use of antibiotics,92 rolled out gradually and evenly over an extended period,
marketing and provision of access to rapid diagnostics,309 and the new discovery and development should be paired
and making outpatient treatment guidelines for with a better and more effective stewardship approach to
standardisation of antibiotic use readily available. protect existing drugs and those that are newly developed.
At the secondary or tertiary level, infection control com- However, if new classes of antibiotics are rolled out too
mittees should be implemented with WHO recom- slowly, millions of people are likely to die because
mendations. Institution champions and teams need to be resistance will arise to all new antibiotics over time. Gaps
chosen. Data collection, analysis, and dissemination in coverage against highly resistant bacteria have already
should be done at primary, secondary, and tertiary levels appeared and there will be larger gaps in the future if more
through pre-prepared software made available by the antibiotics are not produced quickly enough. Furthermore,
national taskforce. For example, in Vietnam despite the discovery of new classes of antibiotics is unlikely to be
legislation supporting antimicrobial resistance contain- steady and even. After all, it might be impossible to
ment, implementation of antimicrobial stewardship at discover and market enough new analogues or classes to
service delivery level became better with technical support keep up with the continuous emergence of antibiotic
from the Vietnam resistance project (VINARES), a private– resistance, especially in Gram-negative bacteria.
public partnership.87 The need for new antibiotics is obvious, but the urgency
varies. In hospitals, especially in intensive care units in
Part 7: Improving the interface between some parts of some countries, there are dangerous levels
academia and the pharmaceutical industry of highly resistant bacteria.318 In other countries, highly
The need for new antibiotics resistant bacteria are less common, especially in the
Antibiotic discovery has stalled, but we do not know how to community.319,320 So the global response to antibiotic
restart the engine. The golden age of antibiotic discovery resistance, in terms of efforts in antibiotic discovery, is
between 1929 and the 1970s saw more than 20 new classes likely to vary with percieved threat.
of antibiotic reach the market.310,311 Since then, only two
new classes have reached the market.312,313 Analogue Why has antibiotic discovery stalled?
development seems to be drying up because the number After the end of the golden age, academic bacteriology
of analogues that can be derived within one class is finite.1 moved away from antibiotic discovery. There was a view
The attitudes of regulators and payers have also that infectious diseases had almost been eliminated and
discouraged development of so-called “me too” products. that industry would market an endless stream of
The net result is that the worldwide antibiotic pipeline for antibiotics. Initially, in the 1970s and 1980s, the pharma-
new antibiotic classes active against highly resistant Gram- ceutical industry did produce a stream of antibiotics. These
negative bacteria is almost dry;314–316 the only novel class in were analogues but not new classes. In retrospect, we
early clinical development has recently been withdrawn.315,317 believe that this was a fundamental mistake, because,
However, the situation for Gram-positives is better because although analogue development is low risk compared with
compounds from two new classes have been marketed in novel class discovery and development, analogues
the past 15 years (linezolid in 2000 and daptomycin in eventually become more difficult to come by. The ingenuity
2006),312,313 and one more is in the pipeline.315 Meanwhile, of the medicinal chemists is overcome by nature, in the
new analogue development and combinations of old form of relentless emergence of new forms of bacterial
antibiotics with, for example β-lactamase inhibitors, are antibiotic resistance. Although resistance arises to first-in-
struggling to keep pace with the relentless emergence of class antibiotics after marketing, a new class of compounds
antibiotic resistant bacteria with new resistance profiles. In widens the opportunities for chemists to create new
our view, if the world is to return to the golden age, it needs analogues that target highly resistant bacteria.
to make new classes of antibiotics and will need up to 20 new Unfortunately, industry, encouraged by academia, made a
classes to reach the market within the next 20–60 years.310 second fundamental mistake, namely to enter genomics
Because resistance arises to all antibiotics, whether they on the grounds that this would lead to many new classes of
are a new class or an analogue of an old class, it could be antibiotics—an endeavour that failed.310 Industry closed
argued that new classes are not needed, only compounds many antibiotic research laboratories. As a result, the
that target highly resistant pathogens, including those world is left with a decreasing stock of effective antibiotics,
resistant to other drugs in its class. an inadequate pipeline of new classes and analogues, a
We believe that, eventually, chemists will run out of broken antibiotic market, a paucity of antibiotic discovery
options if no new classes are marketed because the infrastructure in academia, and insufficient infrastructure
chemical options for variations within one class structure in industry.
are finite. Some Gram-negative bacteria such as NDM-1 A report321 commissioned by the Swedish Government
producing K pneumoniae are already resistant to almost all concluded that the antibiotics market should have rational
antibiotics, including carbapenems; chemists struggle to intervention, the effective life of existing and new
find treatments for some highly resistant bacteria with the antibiotics should be preserved, push–pull incentive

www.thelancet.com/infection Vol 13 December 2013 1079

 The Lancet Infectious Diseases Commission

models should exist, and legal regulatory changes should volume. This is not a desirable situation for new antibiotics.
be made. A high-level conference hosted by the Swedish For example, some countries use antibiotics less well than
Government (figure 7)322,323 concluded that three reasons others, thereby increasing antibiotic resistance, which is
explain the empty pipeline: scientific challenges; regulatory then carried around the world to other countries. On the
requirements, such as needing superiority or non- contrary, delinkage is needed, in which most revenues
inferiority trials; and market failure (ie, absence of should be generated not through sales but through a
appropriate incentives to develop medicines for which use mixed model of income from push funding (such as
will be restricted). The political conclusions made by the through the Innovative Medicines Initiative), pull
EU health ministers, who met after the conference, incentives (such as advance purchase commitments or
resulted in an action plan from the European new market exclusivities), and in-market sales.325
Commission,324 which is being implemented. This meeting Furthermore, models for global access at affordable prices
was followed by a global conference in 2010.325 should be agreed upfront—a fact that emphasises the need
The Pharmaceutical Research and Manufacturers of for new financing and incentives. The advantage of this
America recently asked the FDA for a more flexible approach is that new antibiotics could be put on the shelf
approach to regulation of new antibiotics.326 The European until they are needed, and their use could be restricted,
Medicines Agency has relaxed its guidelines for clinical thereby prolonging their shelf-life. The disadvantage of
antibiotic trials in four ways: first, patients can be enrolled this approach is that the pharmaceutical industry might
in trials after receipt of a dose of previous antibiotic continue to be wary of investment in antibiotic
treatment, making enrolment possible; second, organism- development, although advance purchase commitments
specific rather than disease-specific studies can be done; or new market exclusivities could offset this. The reason
third, small studies can be used to support approval of for industry’s reluctance to invest is because such
antibiotics that treat resistant, critical infections; and restrictions lead to a large reduction in the value of the
fourth, clinical response endpoints can be assessed at test- antibiotic at discovery. The Office of Health Economics330
of-cure visits.327 has estimated that, with restrictions, the net present value
The Infectious Diseases Society of America (IDSA) has of antibiotic to a drug company is minus $50 million,
proposed the 10 x ’20 Initiative, which calls for the whereas a new musculoskeletal drug, without restrictions,
development of ten new, safe, and efficacious systemically is worth $1 billion.327 The Generating Antibiotic Incentives
given antibiotics by 2020, but progress remains elusive.328 Now (GAIN) Act has recently been introduced in the USA
Despite this encouraging attention by policy makers and and allows prolonged exclusivity327 and other provisions.
regulators, some are now questioning whether modern This act could increase the value of antibiotics to
medicine can continue in the future without more effective companies, thereby encouraging companies to enter the
antibiotics.176,329 field; however, it does not address the issue of stewardship.
If a drug is to be used as a last resort, what good are market
The need to improve antibiotic stewardship exclusivity rights?
Under the existing business model, major pharmaceutical The IDSA has suggested331 an alternative way forward
companies presumably market products to maximise that could provide an increased unit price for new
antibiotics and reduce the cost of clinical trials, thus
providing economic incentives for investment while
restricting use. This proposal calls for a new regulatory
approach, the limited population antibiotic drug provision,
through which antibiotics could be approved after small,
low-cost clinical superiority trials that target highly
resistant bacteria that potentially cause lethal disease. The
antibiotic would receive a very narrow label that would
help to protect it from overuse. Although if introduced,
this new regulation could both encourage investment and
restrict use, the high unit price would need to be addressed
Reuters/Pontus Lundahl/Scanpix (Sweden Health)

for global access, especially in developing countries.

The academia–industry interface

The infrastructure of antibiotic discovery in academia and
the pharmaceutical industry has fallen to a dangerously
low level. A prolonged loss of skills in both sectors means
few people have experience of antibiotic discovery,
especially of new classes. If a university or a company has
Figure 7: The Swedish Government, particularly Minister of Health Goran Hagglund, have helped define the no people engaged in antibiotic discovery, it is unlikely to
problem of antibiotic resistance thus far produce new ideas in this area.

1080 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

In academia, emphasis is placed on new ideas. In our be rebuilt, which would be expensive and the responsibility
experience, there is no shortage of new ideas in universities, of industry and governments. Because antibiotic resistance
although there needs to be a critical mass of people in an is believed to be as serious as climate change, governments
institution who think about antibiotic discovery. However, should take appropriate action to address the issue.
translation of these ideas into a marketed antibiotic is Coates333 suggested that the European Commission sets up
difficult, time-consuming, and expensive. Universities and an antibiotic recovery plan, of a similar type to the Marshall
other institutions try hard to address the translation of Plan in 1948. Other major governments (eg, Japan and the
basic research ideas into real products. This issue is USA) should also fund their own plans. Indeed, the USA
sometimes addressed by encouragement of academics to has the elements of such an antibiotic recovery plan,
patent their ideas. Most ideas do not progress much already reasonably well funded, a public–private
beyond this stage because there are not enough partnership mechanism akin to the Marshall Plan, which
entrepreneurs or companies willing to take on high-risk is called the Public Health Emergency Medical Counter-
projects. This gap needs to be filled. measures Enterprise (PHEMCE). This consists of many
Equally relevant is an opportunity to repurpose existing agencies of the US Federal Government, including
compounds and those that were discontinued from National Institute of Allergy and Infectious Diseases,
development with new scientific knowledge. Without Biomedical Advanced Research and Development Agency,
patents and other market exclusivities, few incentives exist and several Department of Defense agencies. This model
to invest in projects to bring back old antibiotics, or to test is in action, as evidenced by various recently announced
new combinations or regimens. multimillion dollar investments. This initiative should be
New ideas for antibiotic discovery also come from small welcomed; however, it has not yet produced the marketed
and medium sized enterprises (SMEs) in the area. Ideas antibiotics that are needed, especially against Gram-
originate from academia, other companies, or from the negatives, perhaps because of insufficient time.
SME itself. Translation of ideas into preclinical antibiotic Additionally, this initiative is probably insufficient to bring
development is something that such enterprises often do enough antibiotics to market for the entire world. A high
well, and some progress into early stage clinical trials. In proportion of loans and grants from governments, say
our view, there is a need for many more SMEs in antibiotic 60%, would be invested as loans in the antibiotic discovery
discovery. A new grant-giving system is needed, and a industry, most of which would be SMEs because novel
new system of loans for SMEs in this area is also ideas increasingly come from smaller companies. The
necessary. Infrastructure to bring these companies and loans would be administered by a bank with the revolving-
academia together is needed at the level of those who are door model.333 Collateral would be provided by government
actually associated with research and development of and interest rates should be very low. However, recipient
antibiotics. This organisation should be at national and companies would need to repay the loans to the
international levels. government. The money could then be lent to another
In the golden era of antibiotic discovery, major company in the antibiotic discovery business. This would
pharmaceutical companies were the home of new ideas, leave 40%, which would be used to fund grants to
development, and marketing. The huge success of universities, SMEs, and major pharmaceutical industries,
antibiotics in shaping modern medicine is mainly a result and pay down debt.
of the efforts of these companies. Unfortunately, those
days are over. Ideas for new drugs, not just new antibiotics, Pharmaceutical industry collaboration
are too few to support major companies. However, phase 3 The Innovative Medicines Initiative, funded by the For more on the Innovative
clinical trials are so expensive (the average cost is European Federation of Pharmaceutical Industries and Medicines Initiative see http://
www.imi.europa.eu/#&panel1-8
$70 million per trial)332 that major companies will usually Associations, is the biggest public–private initiative in
need to be involved in antibiotic drug development. These Europe. It is distinct from the EU Framework Programmes
companies become included in antibiotic development by and aims to accelerate the development of better medicines
in-house efforts, collaborations with academia, buying or for patients. The initiative seeks to improve collaboration
investing in SMEs, or mergers with other large between industry and academia, which has been identified
pharmaceutical companies. Expansion of collaboration as a potential way forward for some years.334 40 projects are
between major pharmaceutical companies and academia in motion that cover many areas of medicine including
is needed. antibiotic resistance (eg, COMBACTE and PreDICT-TB).
Further Innovative Medicines Initiative funds will be
New business models targeted at antibiotic drug discovery in the New Drugs for
Antibiotic recovery plan Bad Bugs programme, if these funds were to be available
The existing resource of antibiotic discovery in academia at early stages of antibiotic discovery that could have a
and industry is too restricted to be effective. An interface substantial effect.
between two organisations, each of which has inadequate The Innovative Medicines Initiative is dominated by
antibiotic discovery, is not going to succeed. The major pharmaceutical partners and is run along the lines
infrastructure of academia and industry needs urgently to of grants from the European Commission, which means

www.thelancet.com/infection Vol 13 December 2013 1081

 The Lancet Infectious Diseases Commission

that deliverables are set at the beginning by the partners. is a good example. The organisation was founded in 2008
The strength of the Innovative Medicines Initiative is its and seeks to provide affordable health care to developing
connection with major pharmaceutical companies, which countries. It is led by the Council of Scientific and
means that drug development is emphasised; its weakness Industrial Research India team with more than
is that it can only fund projects that are already underway, 7500 registered users from more than 130 countries
and might create intellectual property challenges arising around the world who engage in open access drug
from collaborations between public–private partnerships, discovery—making it the largest collaborative group in
academia, and industry. The initiative does not solve the drug discovery in the world. The idea is to discover and
root of the antibiotic discovery problem—too few novel develop drugs with public funding. An example of its
class and analogue candidates in early stage clinical activities is GlaxoSmithKline making 177 potent non-
development. Other collaborative translational research cytotoxic antituberculosis hits publicly available.339 If
models include a new NIH centre in the USA335 and successful, this consortium could help to discover cheap
Cancer Research UK,336 but these are not focused on antibiotics, presumably via the generic market route.
antibiotic discovery. Under present regulatory constraints in developed
countries, it would be difficult to raise sufficient money
Charitable organisations from governments for late-stage clinical trials in, for
The Bill & Melinda Gates Foundation has led the way in example, Gram-negative bacterial ventilator-associated
antibiotic development, especially for tuberculosis. pneumonia, but the regulatory landscape might change to
Bedaquiline, which has been supported by the foundation accommodate much cheaper routes for antibiotics to reach
was marketed in December 2012 for highly resistant the market.
For more on MMV see http:// tuberculosis,337 is the first new drug to be launched for the Another global model is Medicines for Malaria Venture
www.mmv.org/about-us disease in 40 years. This is a model that could also be used (MMV), a not-for-profit public–private partnership, set up
in development of drugs for Gram-negative and Gram- in Switzerland in 1999 with the objective of affordable
positive bacteria. antimalarial drug discovery and development. For
example, MMV in partnership with Sigma-Tau has
Academia–industry collaboration developed dihydroartemisinin–piperaquine (Eurartesim),
The Innovative Medicines Initiative is just one example of which has been granted regulatory approval by the
academia–industry collaboration. Although optimised for European Medicines Agency for the treatment of
development, the predetermined delivery of grants is, in uncomplicated Plasmodium falciparum malaria. This is a
our view, too rigid to stimulate research ideas. Other potential model for the development of new anti-Gram-
models include embedding an SME in academia. This can negatives in collaboration with industry. Although the
work well at the research idea end of the scale, but needs a expense of discovery and development of new antibiotics
robust legal agreement between the SME and the academic might challenge such a model.
For more on the DNDi see http:// institution, and an understanding of industry, which is not The Drugs for Neglected Diseases Initiative (DNDi)
www.dndi.org/about-us/ always present in academia. was founded in 2003 as a not-for-profit drug research
overview-dndi.html
Not-for-profit organisations can also play a part in the and development body. It is a collaboration between
promotion of collaboration. For example, Antibiotic seven global institutions from the public sector: the
Discovery-UK338 was set up in 2012 to bring together Oswaldo Cruz Foundation in Brazil, the Indian Council
universities and SMEs interested in antibiotic discovery. for Medical Research, the Kenya Medical Research
This is simple to set up, by organising meetings of Institute, the Ministry of Health of Malaysia, the
academics and industrial workers who have experience in Pasteur Institute in France, Médecins sans Frontières,
antibiotic discovery, development, and market author- and an international research organisation, the UNDP–
isation. It is a low cost way of starting collaborations that World Bank–WHO Special Programme for Research
can be advantageous when applying for funding from and Training in Tropical Diseases, which acts as a
grant-giving bodies. If other countries were to use this permanent observer. DNDi works on leishmaniasis,
model, representatives of each country could then form— sleeping sickness, Chagas’ disease, malaria, paediatric
eg, Antibiotic Discovery-Europe or America. Antibiotic HIV, and filarial disease in partnership with private
discovery with world coverage would also be a low cost industry, public institutions, academia, and non-
model. Inevitably conflicts of interest will occur, and legal governmental organisations. The initiative has
contracts may be required to prevent industry sponsorship delivered an impressive five new treatments, seven in
from slowing publication of results, restricting the sharing clinical, and seven in preclinical development. The
of findings, or suppressing reporting of adverse drug advantage of this system is the collaboration of the
reactions. public sector with humanitarian and international
Only now is the awareness and urgency of the problem research organisations. The disadvantage is the
of antibiotic resistance reached a level that a new constraints of operating in a low-cost environment,
For more on Open Source Drug
Discovery see http://www.osdd. sustainable global system can be built. For tuberculosis, without obvious financial incentives for large-scale
net/about-us malaria, and leishmaniasis, Open Source Drug Discovery investment from pharmaceutical companies.

1082 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

A new funding model has been developed by the Global Preclinical development would be done by SMEs, and For more on the Global Fund to
Fund to Fight AIDS, Tuberculosis, and Malaria. The fund would aim to mature the drug to be phase 1 ready. A central Fight AIDS, Tuberculosis, and
Malaria see http://www.
operates as a financial instrument, not an implementing facility would need to be set up to provide advice to theglobalfund.org/en/
entity, and supports programmes from national plans. academics and SMEs about preclinical development.
Proposals are assessed by an independent review process. Funding could derive from government loans or grants.
The advantage of such a fund is that the money raised is Phase 1/2 and phase 2 clinical trials would be done by
spent on programmes relevant to partner countries. The SMEs funded by loans and grants. The central facility
fund is not focused on antibiotic resistance, but this could would provide advice about clinical development.
change because some developing countries have a big SMEs would usually falter at the phase 3 stage of
problem already with highly resistant bacterial infections, development, which are often too expensive for such
including tuberculosis. companies. Possible ways forward are to partner with a
Could the World Bank have a role? The World Bank major pharmaceutical company, or with a large financial For more on the World Bank see
provides loans to developing countries for capital institution. Another way is to reduce the cost of trials by http://www.worldbank.org/en/
about
programmes; its goal is to reduce poverty and can provide using microbiological endpoints,6 which can make trials
loans for research and development, which could include up to ten-times cheaper than those using clinical
antibiotic resistance. endpoints.
In summary, regeneration of antibiotic discovery could
Construct an antibiotic discovery programme? be achieved by collaboration between academia and the
On the grounds that one model is unlikely to be enough, pharmaceutical industry. The infrastructure of antibiotic
could we combine many different models and create an discovery in both academia and in industry is dangerously
effective global antibiotic discovery programme? weakened and needs to be rebuilt.
Countries could set up not-for-profit overarching
networks of people involved in antibiotic discovery. These Part 8: Beyond antibiotics—alternative strategies
networks would then form the basis of a global network for prevention and treatment
and could give rise to an open source for antibiotic Antibiotics today
discovery. Collaborations with major partners such as Most antibiotic chemical scaffolds in present clinical use
countries and institutions could be modelled on the were discovered more than 50 years ago. These discoveries
DNDi. Funding could come from inclusion of antibiotic mainly came as a result of mining the bioactive chemicals
discovery by charitable organisations such as the Gates produced by soil bacteria. The large effect of these new
Foundation and by the Global Fund and World Bank. antibiotics in treating infections and enabling new medical
Changes to the regulation of antibiotic development for procedures was tempered somewhat early in their use by
highly resistant bacterial pathogens using the IDSA the emergence of resistant strains of pathogens and
suggestion of the limited population antibiotic drugs eventually their spread, along with their genes, across the
could be used. Alternatively, regulations could be globe.4 Later, many antibiotic scaffolds derived from
changed towards those that already exist for annual microbial secondary metabolism were chemically modified
influenza vaccines, in which postmarketing safety to improve efficacy and circumvent resistance, leading to
surveillance would be mandatory. Cofunding through regular availability of plentiful new medicines to address
the Innovative Medicines Initiative is an attractive emerging challenges from increasingly multidrug-
solution, but will only work when a financier has been resistant pathogens.
found for the 50% required by the private sector. In The ability of the pharmaceutical sector to reliably deliver
addition, a loan system, of the type used in the Marshall new drugs to the clinic to address the resistance-mediated
Plan, would enable a large number of phase 3 clinical obsolescence of old medicines is very limited in the 21st
trials, the most expensive part of antibiotic development, century. A general paucity of investment and activity in
to be done. Loans are preferable to grants in the long- antibiotic drug discovery exists, and consequently few new
term, because once paid back money can be reinvested drugs are on the horizon.340 This fact has frankly taken the
into drug discovery. public, clinicians, regulators, and politicians by surprise.
A grand challenge in antibiotic discovery would be one Where will the new antibiotics come from that are needed
place to start. This idea is based upon the grand challenges to address the growing problem of resistance? Both
used by the Gates Foundation. Grants, be they from conventional and non-conventional approaches are needed
government or charity, would be given with a low entry to address this pressing clinical problem.341
threshold to academia and SMEs that had new ideas of
antibiotic discovery. There may be an insufficient number New antibiotics
of SMEs to cope with the need for new compounds— Once dismissed as old thinking, a renaissance in natural
academics could be encouraged to set up their own product discovery is needed. We have not exhausted the
companies and SMEs could be encouraged to embed in chemical space of traditional single molecules with potent
universities to foster collaboration with legal agreements antibiotic activity and drug potential. However, the days of
to protect both parties. highly potent, non-toxic, broad-spectrum, and inexpensive

www.thelancet.com/infection Vol 13 December 2013 1083

 The Lancet Infectious Diseases Commission

drugs might be over. These were the criteria that and increased clinical need spurred by the emergence of
traditionally benchmarked antibiotic drug discovery and MRSA as an important pathogen resulted in re-
are increasingly difficult to achieve. In fact, the desire for examination of daptomycin by Cubist, a medium sized
broad-spectrum agents, active against most Gram-positive pharmaceutical company, and subsequent clinical approval
and Gram-negative pathogens, has selected for drugs with of daptomycin in 2003. Sales of the drug are on track to top
the capacity for serious deleterious effects on the human $1 billion per year in the near future, showing that even
microbiota. The result has been unwanted side-effects narrow-spectrum natural product agents have the potential
and, in the emergence of antibiotic associated colitis due to to provide excellent return on investment. Indeed,
C difficile, even increased morbidity and mortality.342 The measures such as the GAIN Act, signed into law recently
clinical reasoning driving the search for broad-spectrum by the US Government, provide for additional market
drugs is the fact that very often, the microorganisms exclusivity and a favourable regulatory framework for new
causing the infection are unknown to the clinician at the antibiotic drugs.349
time of presentation. Therefore physicians opt for Therefore, with the appropriate chemical matter, clever
empirical treatment to cover all the bases, leading to a screens that avoid prevailing chemical scaffolds, and a
desire for broad-spectrum drugs. A new focus on very willingness to forego broad-spectrum coverage for a more
narrow spectrum or even organism-specific antibiotics limited field of susceptible organisms, there is great reason
accompanied by advances in molecular diagnostics to to be optimistic that traditional single agent antibiotics can
guide infectious disease specialists in drug deployment continue to be identified and brought to market. However,
could provide a new, more selective antibiotic arsenal. there are also grounds to be optimistic that alternative
However, the discovery of such drugs will need new approaches to traditional antibiotics can offer alternative
approaches and chemicals in step with innovation in strategies for new medicines. These include, but are not
diagnostics and changes to the regulatory frameworks limited to, antibiotic adjuvants, antivirulence strategies,
governing new antibiotic approval.343 and biologicals.
The chemical matter and drug discovery strategy
dominant in the antibiotic sector over the past 25 years has Adjuvants
been focused on synthetic chemical libraries and in-vitro Compounds that potentiate the activity of antibiotics,
target-based approaches. Although this strategy has antibiotic adjuvants, can increase the efficacy of drugs in
enjoyed success in various therapeutic areas, its use in both antibiotic susceptible and resistant pathogens.350 The
antibiotic discovery is a well documented failure.344 Modern combination of antibiotics, often empirically to cover the
target-based drug discovery seems to have missed the spectrum of possible infections and to avoid possible
mark because of an incomplete understanding of the resistance, has been a mainstay of infectious disease
bacterial targets selected and of the chemistry needed for medicine for decades. Indeed, the first-line treatment of
success in the development of antibacterial drugs. some bacterial diseases, such as tuberculosis, requires
However, natural products, in particular from microbial multidrug treatment, as do common treatments for cancer
sources, have been an outstanding source of antibiotic and viral diseases like hepatitis and HIV/AIDS. However,
chemical scaffolds. More than 25 000 such molecules have combination of antibiotics with non-antibiotic compounds
already been discovered and the genome sequences of with adjuvant activity is a relatively novel strategy that can
producing organisms have provided a glimpse of a huge reveal unexpected biology and overcome resistance.
potential for new compounds.345,346 Reinvestigation of this Antibiotic adjuvants can take the form of compounds
bioactive chemical wealth should prove useful, especially that directly target resistance mechanisms and compounds
with cell-based screens on either multidrug-resistant that exploit chemical genetic interactions that potentiate
organisms to avoid identification of known drug scaffolds antibiotic activity. Directly targeting resistance mechanisms
or strains engineered to favour discovery of desired is a clinically proven strategy that extends the spectrum of
targets—for example using antisense RNA strategies to antibiotics by inhibiting resistance and rescuing antibiotic
deplete key enzymes or pathways thereby sensitising cells activity. Various inhibitors of Ser-β-lactamases in
to compounds affecting privileged targets.347 combination with penicillins have been in clinical use for
The success of the lipopeptide daptomycin (marketed as decades.351 Important among these is a combination of
Cubicin), is an instructive lesson to emphasise some of the amoxicillin with the β-lactamase inhibitor clavulanic acid,
above points.348 Daptomycin is a natural product produced a natural product produced by Streptomyces calvuligerus,
by the soil bacterium Streptomyces roseosporus first marketed as Augmentin (GlaxoSmithKline). Other similar
discovered in the 1980s at Eli Lilly. Initial exploration of its combinations in clinical use include ticarcillin with
promise as a new antibiotic was halted because of muscle clavulanic acid (Timentin; GlaxoSmithKline), piperacillin
toxic effects. Furthermore, daptomycin had a narrow with tazobactam (Zosyn; Pfizer) and ampicillin with
spectrum, effective only against selected Gram-positive sulbactam (Unasyn; Pfizer). Experimentally, clavulanic
pathogens, a deal-breaker for conventional antibiotic acid inhibits the intrinsic β-lactamase that renders
development strategies. Nevertheless creative re- Mycobacterium tuberculosis impervious to β-lactam
examination of dosing obviated the muscle toxic effects antibiotics and thus a combination with meropenem has

1084 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

excellent killing activity against this deadly pathogen prevent biofilms from forming, disrupt them, or otherwise
offering a new treatment strategy.352 block their antibiotic insensitivity have been reported.359,360
The β-lactamase inhibitor combinations in current Some of these are quite potent in their effects in vitro,
clinical use have little effect on strains producing AmpC or suggesting that the approach has merit for clinical
ESBLs (figure 8). To address this clinical need, the novel assessment.
β-lactamase inhibitor avibactam in combination with
cephalosporins (ceftaroline and ceftazidime) is in late- Antivirulence strategies
stage clinical trials and is showing great promise.353 The One of the reasons that antibiotic resistance is such a
notion of directly inhibiting resistance elements, thereby challenge is that the evolutionary bottleneck of cell death is
rescuing antibiotic activity, is not confined to the β-lactam so stringent, and the numbers of bacteria in one infection
antibiotics. Efforts to block aminoglycoside-inactivating are so large that even rare mutational events can result in
enzymes354 and antibiotic efflux pumps355 have also proven resistant mutants that break through. Coupled with the
effective in vitro, though none have progressed to clinical ability to receive DNA horizontally from other species or
trials yet. even genera, the massive adapt or die selective pressure
Antibiotic adjuvants are not restricted to inhibitors of steers evolution towards the selection of ever more
resistance. In a screen of off-patent drugs for potentiators resistant pathogens. An alternative notion is not to kill
of the tetracycline analogue minocycline against E coli, bacteria, but rather impair their ability to establish an
S aureus, and P aeruginosa, several non-antibiotic infection in the first place. These antivirulence strategies
compounds were identified with adjuvant properties. One have been aided by genetic screens to identify virulence
of these was loperamide (Immodium; McNeil Consumer determinants for a number of pathogens.361 In principle,
Healthcare).356 This compound is a μ-opioid receptor such virulence targets are less prone to selection for
agonist in people. In Gram-negative bacteria (that do not resistance to antivirulence agents, and mutants that do
have this receptor), loperamide disrupts the electrical occur are much more likely to be impaired in their ability
component (Δ
) of the proton motive force across the to cause infection. To the extent that some antivirulence
bacterial membrane. The bacteria respond by increasing approaches will place a selective pressure on pathogens,
the pH gradient (ΔpH) to maintain the ATP-generating one clearly needs to be cautious about the question of drug
capacity of the proton motive force. It is this pH gradient resistance and virulence. Nevertheless, this approach will
that enables tetracycline to enter the cell. Therefore benefit from the fact that antivirulence compounds tend to
loperamide increases antibiotic influx, even in otherwise be narrow in spectrum and this will restrict any selective
resistant cells. The combination was highly effective in pressures for drug resistance. Here again, the benefit of
vitro and in animal models of infection, providing a proof- species-specific treatment defines an imperative for a new
of-principle of the adjuvant approach and screening to generation of antibiotics.
identify non-obvious drug combinations. Similar screens Among the more promising antivirulence targets are
in pathogenic yeast uncovered unexpected cell-killing quorum sensing, type 3 secretion systems, toxins, and the
synergy between the fungistatic antifungal drug biosynthesis of glycolipid surface structures. Quorum
fluconazole and the selective serotonin reuptake inhibitor sensing refers to the production of diffusible small
sertraline (Zoloft; Pfizer).357 molecules by bacteria that act as autoinducers of various
In recently published work,358 the approach identified a cellular factors at high cell densities. These compounds,
new combination of the cephalosporin cefuroxime and the such as homoserine lactones produced by Gram-negative
antiplatelet aggregation drug ticlopidine (Ticlid; Roche).
Here an unexpected inhibition of TarO, an enzyme
essential for early stage wall teichoic acid biosynthesis, by
ticlopidine was recorded. This inhibition results in
particularly powerful synergy with β-lactam antibiotics that
target PBP2 in MRSA, which otherwise is insensitive to
cefuroxime. Thus, the combination achieves both
suppression of antibiotic resistance and narrow pathogen
selectivity, both key attributes of 21st century antibiotic
medicines.
The antibiotic adjuvant idea can also be applied to
Animate4.Com/Science Photo Libary

overcome bacterial functions that are commonly associated

with drug resistance. In particular, bacteria can adhere to
surfaces such as medical devices and epithelial cells, where
they continue to grow but become insensitive to antibiotics.
These biofilms are implicated in numerous bacterial
infections and are often the reason that antibiotic treatment
fails. Several efforts to combine antibiotics with agents that Figure 8: Potential targets of new drugs include resistance enzymes such as ESBLs

www.thelancet.com/infection Vol 13 December 2013 1085

 The Lancet Infectious Diseases Commission

pathogens and small peptides by Gram-positive organisms, of the causative agent of disease. In addition to the
trigger various responses including the formation of challenge of diagnosis, to be fully effective such agents
biofilms. Inhibition of quorum sensing disrupts this might need to be provided before infection becomes well
process. Several such inhibitory compounds have been established. In some cases—for example, where C difficile
identified that block virulence and quorum sensing in is a known and serious risk—the argument for prophylaxis
P aeruginosa362 and S aureus,363,364 for example. will be relatively straightforward, but in other cases, it will
Bacterial colonisation of tissue is essential to the not be. Further, these challenges will make clinical trial
establishment of infection and interference with this design especially complex. It is nevertheless imaginable
process is a high-quality target for antivirulence that the first implementation of such drugs may be in
compounds. Among the first small molecules shown to combination with conventional antibiotics.
prevent virulence was virstatin.365 This molecule blocks the
activity of the V cholerae transcription factor ToxT that Biological therapeutics
controls the expression of TCP, a pillus protein that is With growing concern over the failing antibiotic drug-
essential for colonisation of the intestine. Similarly type 3 discovery pipeline, there is a great deal of interest and
secretion systems are extracellular proteins that facilitate efforts in progress to investigate alternatives to small
bacterial binding to cells and injection of virulence proteins molecule drugs. Where biological medical products have
into the cytosol. Several small molecule inhibitors of type 3 had remarkable effect on modern medicine, this class of
secretion systems have been identified including therapeutics is increasingly touted as a viable alternative
salicylidene acylhydrazides that are effective against several to antibiotics. Indeed, there has been remarkable success
pathogens including Chlamydia spp, E coli, S enterica, and in the area of biologicals (eg, insulin, erythropoietin,
Shigella flexneri.366 monoclonal antibodies, and other engineered receptor
Many pathogens, such as C difficile, V cholerae, Bordetella constructs) to provide therapeutic options in which there
pertussis, E coli, and Bacillus anthracis, secrete extracellular were no or few effective alternatives in other areas of
toxins that contribute greatly to virulence. These toxins are medicine. Thus, although biologicals might not form the
often enzymes and have been targeted by screening mainstay of future strategies to deal with multidrug
campaigns to identify inhibitors that attenuate virulence. resistance, substantial interest exists in various areas
For example, a series of small molecules have been recently including monoclonal antibodies, modifiers of the
reported that block ADP-ribosyltransferases that are immune system, and age-old phage treatment. We deem
known extracellular bacterial toxins associated with the first two of these approaches to be particularly
cholera, diphtheria, and pertussis.367 exciting new directions for therapeutics to treat bacterial
Disruption of the production of glycolipid surface infection. On the whole, these approaches are not being
structures, such as lipopolysaccharide in Gram-negative thought of in isolation of antibiotic treatments but as
pathogens and lipoteichoic acids in Gram-positive adjunctive therapies to increase the efficacy of antibiotic
bacteria, are promising antivirulence strategies. In terms treatment. Efforts on biologicals in the form of
of the former, the essential in-vitro early steps of monoclonal antibodies and immune modulators are
lipopolysaccharide assembly—namely lipid A and particularly innovative directions that will benefit from
3-deoxy-D-manno-oct-2-ulosonic acid biosynthesis—have new understanding of the host–pathogen interactions.
long been indicated as valid targets. However, late steps in These are more long-term solutions to the challenge of
lipopolysaccharide synthesis are dispensable in vitro, but antibiotic resistance, but they are driven by the urgent
are associated with substantial reductions in virulence and growing need for new treatments.
and increased antibiotic sensitivity. Data from proof-of- The past couple of decades have seen many successful
principle efforts to find late-step inhibitors368,369 suggest monoclonal antibody-based products introduced in the
that these are targets of high potential. Lipoteichoic acid areas of rheumatology and oncology. Collectively, these
biosynthesis, a polyanionic glycolipid common to most efforts have made the discovery, manufacture, and
Gram-positive pathogens has also been the target of clinical development for monoclonal antibodies a mature
recent small molecule screening efforts that have pharmaceutical approach for consideration in an
generated interesting new leads.370 Additionally, data from increasing variety of therapeutic areas. This general
feasibility studies in M tuberculosis have suggested that approach was formerly the exclusive domain of treat-
the synthesis of acyltrehalose-containing glycolipids is ments for infection. Passive vaccination, the provision of
crucial to host invasion,371 setting the stage for readymade antibodies to treat infectious disease, was
antivirulence drug discovery efforts aimed at the cell developed by the likes of Paul Ehrlich and Emil Behring
surface of this organism. in the late 19th century and was the standard of care until
Although antivirulence strategies are expanding the it was supplanted by antibiotic treatment. However,
target base for new antibiotics, several challenges are today, only one anti-infective monoclonal antibody is on
unique to this approach. First, antivirulence agents are the market—palivizumab for the prevention of
probably highly specific to individual pathogens and there respiratory syncytial virus infections in high-risk
will be a requirement for great confidence in identification neonates.372

1086 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

Modern discovery and development efforts in faecal transplants) for the treatment of severe
monoclonal antibodies for bacterial diseases have been gastrointestinal disease caused by C difficile infection has
largely focused on bacterial toxins or on cell surface promise. A recent systematic review and meta-analysis379
structures.373 The protective antigen of B anthracis, for of available clinical data reported remarkable resolution
example, has been a popular target as are the toxins of of disease in most cases, but cautioned that further
C difficile. Lipid A and lipoteichoic acid have likewise randomised controlled trails and long-term follow-up are
been targets for monoclonal antibody treatment now needed to confirm the usefulness of the approach.
development, as have surface proteins and carbohydrates Another creative approach that seeks to prevent C difficile
from various important pathogens. Although it is overgrowth and infection resulting from antibiotic use is
relatively early days for these approaches, a consensus is the oral administration of antibiotic resistance enzymes.
emerging that success in animal models of infection is The biotechnology company Synthetic Biologics is
not always predictive of clinical success.374 Furthermore, developing a treatment of β-lactamases that when taken
combinations of monoclonal antibodies, perhaps in orally will degrade β-lactam antibiotics in the intestine
addition to antibiotic treatment, might be a preferred that otherwise could degrade the normal microbiome and
approach. facilitate C difficile infection.
Modulation of host immunity is another exciting and Finally, vaccines remain powerful agents for disease
emerging approach for the application of biological prevention. Their use obviates the need for antibiotic by
medicine products to treat bacterial infections. Here, the preventing infection in the first place, and they have been
natural defence mechanisms of the host are targeted for in clinical use for decades. This is an area in which
enhancement to derive therapeutic benefit. Immune epidemiological studies that identify key circulating
modulators under investigation include host defence pathogens are essential to continuously revise vaccine
regulator peptides and agonists of the innate immune formulations. The effectiveness of the new seven-valent
system—eg, Toll-like receptors and NOD-like pneumococcal conjugate vaccine in reducing disease
receptors.375 Innate defence regulator peptides are a burden and consequent unintended selection for the rise
particularly interesting subset of these modulators, and of serotypes that are not covered by the vaccine shows the
are often cationic peptides (10–50 aminoacids) with a importance of vaccination and continuous monitoring of
high proportion of hydrophobic residues that impart target organisms.380 Vaccines directed against organisms
amphiphilic physical properties, are generally for which antibiotic resistance is especially challenging is a
membrane-active, and have antimicrobial activity in strategy that could have great clinical effect. Recent efforts
vitro.376 Nevertheless, these peptides often have only to develop vaccines for important drug resistant pathogens
weak antimicrobial activity under host physiological such as MRSA381 and M tuberculosis343,382 are promising, but
conditions and have convincing effects in modulating have yet to generate viable candidates for late stage clinical
the host immune response. The clinical development of trials.
such molecules to treat bacterial infections has
commonly focused on the antimicrobial activities with Part 9: Call to action
subsequent recognition during development of the Measuring the extent of the problem and its
immunomodulatory properties.377 Thus, these seem to consequences
be combination treatments with immune and antibiotic The generation of reliable, relevant, and up-to-date
activities in one medicine. information will be essential to respond to the negative
Also in this category of immune modulating biologicals effects of antibiotic resistance on public health. The poor
are probiotics. A mounting recognition of the role of host understanding of the unique features and risks of
microbiota in shaping and modulating our immune antibiotic resistance is an important cause for the global
systems has led to the therapeutic application of probiotics complacency paving the way for the present crisis. Few
for various diseases, most notably the treatment of C studies have been done on the magnitude of the burden of
difficile.378 Such treatments have intuitive appeal to patients antibiotic resistance and its contributions to excess
and are growing in popularity. Indeed, the science of mortality to convince policy makers of the need to react.
probiotics remains a mysterious one at present, but the Although antibiotic resistance is undermining the effective
potential is clearly there for this approach and for immune treatment of many important bacterial diseases with high
modulation treatments generally. The specialty is well mortality, especially in LMICs, it lacks the profile of HIV,
positioned for transformational insights from break- tuberculosis, and malaria. Clear information on the health
throughs in mechanistic understanding of how to control and economic burden of antibiotic resistance is urgently
the immune system in treatment. needed to make this complex problem tangible to policy
Other antibiotic alternatives that in particular have makers. Studies in the EU383 and USA36 provide some data,
potential effect on C difficile infections are the use of but need to be scaled up worldwide, and the results from
microbiome transplantation and the therapeutic such data collection and analysis should be communicated
application of antibiotic resistance proteins. The trans- to decision makers, prescribers, dispensers, and the
plantation of intestinal microbiomes (also known as general public. Likewise, a global surveillance system for

www.thelancet.com/infection Vol 13 December 2013 1087

 The Lancet Infectious Diseases Commission

For more on ADMER see http:// antibiotic resistance, including outbreak reporting and an the Danish supported Antibiotic Drug Use Monitoring
admerproject.org/ early warning system to detect new resistance mechanisms and Evaluation of Resistance (ADMER) in Ghana. Rapid
and their global spread, is still lacking, despite the obvious point-of-care diagnostic tests providing information on the
need of such a system and several proposals for potential pathogen and its susceptibility to antibiotics have
models.384,385 Absence of essential local epidemiological enormous potential to minimise inappropriate antibiotic
data also leads to delayed or suboptimum revisions of use and to increase patients’ safety. Additionally, cost
treatment guidelines, thereby driving the vicious circle of savings from diagnostic testing might motivate their
injudicious empirical use of antibiotics by prescribers development. So far, efforts to develop such technology
based on anecdotal evidence or experience without a firm have been fragmented, with a lack of corporate learning
evidence base. One large hurdle is the insufficient and inadequate identification and sharing of multi-
laboratory capacity in LMICs, which need to be disciplinary expertise. The present diagnostic development
strengthened. Meanwhile, sentinel population-based landscape is scattered, leading to tests of suboptimum
studies are needed to get a rapid global assessment of the overall quality and insufficient clinical relevance.
scale of antibiotic resistance in key pathogens. Likewise, Challenges exist at every step from ideas for rapid
data for antibiotic use in human medicine, veterinary diagnostic tests to uptake by health-care systems.
medicine, agriculture, and horticulture need to be A multidisciplinary, multistakeholder, multiregional,
monitored globally and regionally. All countries should set long-term platform should be created with the task of
up systems for the monitoring of antibiotic use and developing a living road map for diagnostic development,
comparable units of measurements should be agreed and identifying the clinical need and associated (technical)
implemented. requirements in view of market and technology forecasts.
The road map should identify the present and future
Regulation, rational use, and infection control in clinical needs and diagnostic requirements, which would
human medicine enable test developers to bring together appropriate
Antibiotics are different from all other medicines in that technologies to provide a clinically useful answer. In
the effects of their use extend way beyond individual addition to technology developers (commercial and
patients. The societal effects of antibiotic use justifies that academic), experts in clinical, statistical, regulatory,
they should belong to a special regulatory category. behaviour change, and social and health economic
Antibiotic use should be strictly monitored and legislation sciences, the platform should include the end users—
to prevent over-the-counter sales without a prescription clinical microbiologists, infectious diseases specialists,
enforced, unless this would cause an unacceptable access and other health-care providers—who will base treatment
problem (eg, in rural areas). Financial incentives to decision on the results of these tests. Factors that affect
prescribers and dispensers leading to irrational use need the introduction of new diagnostic technologies and
to be removed. Regular revision of standard treatment delivery into health systems should be identified, and
guidelines into clear, simple, updated, evidence-based, barriers to acceptance of rapid diagnostic tests addressed.
locally relevant, and accessible documents is essential. End-users should develop diagnostic algorithms for
Motivational measures include pay-for-performance policy, major infectious diseases where rapid diagnostics could
the audit-feedback mechanism on antibiotic prescribing be used to target patients who really need antibiotics. The
rates of individual prescribers, and public disclosure on road map should be developed in three steps: first, reach
antibiotic prescribing rates of health-care facilities. To a consensus about a set of clinical needs and the
reduce patients’ perceived need for antibiotics and reduce technologies needed to satisfy those needs; second,
demand and change social norms, well designed and provide a mechanism to help forecast technology
contextualised educational campaigns should be done. developments; and third, provide a framework to help
Implementation of basic hygienic routines is still largely plan and coordinate technology development. The
absent in many hospitals and health-care settings creation of target product profiles for novel diagnostic
worldwide. Cultural and other barriers for their tests would provide useful guidance for industry. Product
implementation need much more study. Infection control development partnerships like FIND or PATH could
interventions need to be reassessed and improved in an potentially embrace this area of work.
era with rapid transmission of multidrug-resistant bacteria
and mobile antibiotic resistance genes. The animal sector and the environment
Antibiotics and antibiotic resistance needs to be seen from
The need for rapid diagnostics an ecological and environmental perspective. Strong local
A fundamental obstacle in the management of antibiotic and global partnerships in which policy makers, academia,
resistance in LMICs is the inadequate capacity and and professionals from all sectors work together to
infrastructure to do basic microbiological laboratory improve present systems are needed to ensure multifaceted
analyses. These deficiencies need to be addressed—eg, by action, sharing of experiences, transparency, and con-
mechanisms similar to the World Bank supported East tinuous improvement. Unnecessary antibiotic use in all
Africa Public Health Laboratory Networking Project386 and sectors needs to be removed and the spread to the

1088 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

environment minimised. For the animal sector and the substantial push funding for antibiotic discovery and
environment, we propose the following policy actions. development, incentives are needed to secure the rational
Use of antibiotics, which is unnecessary (eg, for growth use of new antibiotics and prevent them from being
promotion) or where alternatives exist (eg, routine heavily marketed. Financing models that delink research
prevention) should be phased out. The international and development investment from revenue returns need
organisations WHO, OIE, and FAO should provide a clear to be developed and piloted. By such delinkage, the return
definition of “unnecessary routine prevention”. on investment for pharmaceutical firms should be
Governments across the globe should then revise existing divorced from volume-based market sales of antibiotics
legislation or draft new legislation accordingly. The and access to new antibiotics will need to be controlled
transition should be supported by extension services through regulation, procurement, and distribution. Such
providing advice on alternatives and on how to tackle approaches should empower consumers and providers
potential negative animal health problems. Resources alike to tackle antibiotic resistance as well as ensure that
should be allocated for a dialogue between all stakeholders those in need benefit from affordable, effective antibiotics.
and for monitoring of compliance with the new legislation. Intensive research efforts need to be directed to retain
WHO, OIE, and FAO should monitor progress the efficacy of our present antibiotics by substantial
All countries should monitor use of antimicrobials for investment in antibiotic adjuvants. Alongside antibiotic
animals in a comparable way. Guidance on collection of research, increased investments need to be made in
data and units of measurement should be agreed on by alternatives to antibiotics. The goal is to encourage
WHO, OIE, and FAO. Governments should set up systems researchers and biotech to reconsider the reliance on
for data collection and analysis of antibiotic use by species antibiotics and to explore other means of controlling
and production type. The pharmaceutical industry, farmers, bacterial infection, such as virulence blockers, monoclonal
and veterinarians should contribute by providing access to antibodies, immune modulators, and bacterial vaccines.
data. WHO, OIE, and FAO should monitor progress jointly.
The need for antimicrobials in animals should be Governmental coordination and action
reduced. WHO, OIE, and FAO, governments, and key The ultimate responsibility for the provision of equitable
stakeholders (pharmaceutical industry, farmers, and affordable access to effective antibiotics for those in
veterinarians, academics, among others) should work need lies with national governments. The consequences of
together to identify and implement incentives for antibiotic resistance reach far beyond the human health
development of health-orientated systems for rearing of sector and thus no one governmental ministry or agency
animals for food production and for development and can be held solely responsible. National task forces with a
implementation of other disease preventive measures. broad intersectoral coordinating role, including all relevant
Additionally, all key stakeholders should commit to stakeholders, with governmental mandate, are needed.
prudent and rational use of antimicrobials. The Task forces should produce annual action plans and
environmental release of antibiotics from all sectors needs milestones in different areas, such as surveillance,
to be monitored and controlled. Strategies need to identify regulation, treatment guidelines, infection control,
and focus control on hot spots for horizontal resistance education, and awareness raising. Health professionals
gene transfer such as wastewater treatment facilities. and university academics have an important catalytic role
and should engage with community leaders, civil society
New antibiotics and alternative strategies organisations, and students.
The rapid pandemic spread of multiresistant bacteria and
the paucity of new effective antibiotics is placing patients’ Global governance
safety at risk worldwide. The infrastructure of antibiotic The serious threat to public health caused by the rapid loss
discovery both in academia and in industry is at a of antibiotic effectiveness calls for concerted global action.
dangerously low level and needs to be rebuilt. A new We advocate for a health systems thinking approach in the
sustainable global model for the discovery, development, efforts to contain antibiotic resistance. Treatment for
and distribution of antibiotics needs be developed in which bacterial infections is a fundamental prerequisite for
the private and public sectors work in partnership and the fulfilment of health related Millennium Development
large scientific bottlenecks for discovery of antibiotics with Goals and access to effective antibiotics should be included
new mechanisms of action are solved in collaboration in the discussions of the post-2015 Developments Goals.
between academia, SMEs, and major pharmaceutical We call for a coalition of governments with a strong
companies. In the EU, substantial public funding for representation from LMICs that will work with WHO,
antibiotic research has been put into the Innovative UNICEF, UN Development Programme, other UN
Medicines Initiative, and the Joint Programming Initative agencies, other international bodies, science academies, For more on the Initative on
on Antimicrobial Resistance is strengthening collaboration development aid agencies, philanthropists, and civil Antimicrobial Resistance see
http://www.jpiamr.eu/
within the academic sectors in 18 of the EU member states. society organisations to develop a global plan to tackle the organisation/
These important initiatives need to be moved to structured antibiotic crisis and share responsibilities for its
sustainable global collaboration. Beyond the much needed implementation. This governmental coalition should also

www.thelancet.com/infection Vol 13 December 2013 1089

 The Lancet Infectious Diseases Commission

initiate processes to create global funding mechanisms for 15 Viswanathan R, Singh AK, Ghosh C, Dasgupta S, Mukherjee S,
needs driven research and development of antibiotics, Basu S. Profile of neonatal septicaemia at a district-level sick newborn
care unit. J Health Popul Nutr 2012; 30: 41–48.
alternative treatments, diagnostics, and vaccines. A global 16 Saleem AF, Ahmed I, Mir F, Ali SR, Zaidi AK. Pan-resistant
code of conduct should be developed for antibiotic use and Acinetobacter infection in neonates in Karachi, Pakistan.
implementation of strategies against antibiotic resistance. J Infect Dev Ctries 2010; 4: 30–37.
17 Perry JD, Naqvi SH, Mirza IA, et al. Prevalence of faecal carriage of
Contributors Enterobacteriaceae with NDM-1 carbapenemase at military hospitals
A list of the contributors to each section can be found in the panel on in Pakistan, and evaluation of two chromogenic media.
page 1 of the Commission. Part 1: RL did the literature search and wrote J Antimicrob Chemother 2011; 66: 2288–94.
the section, AKMZ, AD, and CW contributed to the writing and read the 18 Khan E, Ejaz M, Zafar A, et al. Increased isolation of ESBL producing
final paper. Part 2: all authors contributed to the reference search for the Klebsiella pneumoniae with emergence of carbapenem resistant
piece, writing, revisions, and final approval. Part 5: ADS drafted the isolates in Pakistan: report from a tertiary care hospital.
original piece; all authors participated in all subsequent phases and J Pak Med Assoc 2010; 60: 186–90.
approved the final manuscript. Part 6: FNQ and FM contributed equally. 19 Perovic O, Singh-Moodley A, Smith M, Lowman W. Antimicrobial
resistance of Klebsiella pneumoniae and Staphylococcus aureus at sentinel
Conflicts of interest sites in South Africa. 52nd interscience conference on antimicrobial
AC holds an European Commission FP7 grant(BacAttack) and an agents and chemotherapy; San Francisco; Sept 9–12, 2012.
Innovative Medicines Initiative grant (Predict TB) and is the founder, 20 Lowman W, Sriruttan C, Nana T, et al. NDM-1 has arrived: first report
director, chief scientific officer, shareholder of Helperby Therapeutics Ltd of a carbapenem resistance mechanism in South Africa. S Afr Med J
which makes new antibiotics. All other authors declare no conflicts of 2011; 101: 873–75.
interest. IMG has received consultancy and lecture and as president of the 21 NICD-NHLS. Update on carbapenemase-producing
International Society of Chemotherapy frequently requests meeting Enterobacteriaceae. NICD Communicable Diseases Communiqué 2012;
support from various companies involved in the development of 12: 5–6.
diagnostics and antibiotics for MRSA. 22 Hollyway K, Mathai E, Sorensen T, Gray T. Community-based
surveillance of antimicrobial use and resistance in
Acknowledgments
resources-constrained settings: report on five pilot projects. Geneva:
FNQ and FM received research training support from the National World Health Organization, 2009.
Institute of Health’s Fogarty International Center (1 D43 TW007585-01).
23 Datta S, Wattal C, Goel N, Oberoi JK, Raveendran R, Prasad KJ. A ten
ADS received funding from the Robert Wood Johnson Foundation year analysis of multi-drug resistant blood stream infections caused
Investigator Award in Health Policy Research and a SIDA-supported grant by Escherichia coli and Klebsiella pneumoniae in a tertiary care hospital.
to ReAct—Action in Antibiotic Resistance. Indian J Med Res 2012; 135: 907–12.
References 24 Ko KS, Suh JY, Kwon KT, et al. High rates of resistance to colistin and
1 Centers for Disease Control and Prevention (CDC). Vital signs: polymyxin B in subgroups of Acinetobacter baumannii isolates from
carbapenem-resistant Enterobacteriaceae. Korea. J Antimicrob Chemother 2007; 60: 1163–67.
MMWR Morb Mortal Wkly Rep 2013; 62: 165–70. 25 Capone A, Giannella M, Fortini D, et al, for the SEERBIO-GRAB
2 Laxminarayan R, Heymann DL. Challenges of drug resistance in the network. High rate of colistin resistance among patients with
developing world. BMJ 2012; 344: e1567. carbapenem-resistant Klebsiella pneumoniae infection accounts for
an excess of mortality. Clin Microbiol Infect 2013; 19: 23–30.
3 Abraham EP, Chain E. An enzyme from bacteria able to destroy
penicillin. 1940. Rev Infect Dis 1988; 10: 677–78. 26 Kontopidou F, Plachouras D, Papadomichelakis E, et al. Colonization
and infection by colistin-resistant Gram-negative bacteria in a cohort
4 Davies J, Davies D. Origins and evolution of antibiotic resistance.
of critically ill patients. Clin Microbiol Infect 2011; 17: 9–11.
Microbiol Mol Biol Rev 2010; 74: 417–33.
27 Antoniadou A, Kontopidou F, Poulakou G, et al. Colistin-resistant
5 Holmes KK, Johnson DW, Floyd TM. Studies of venereal disease. I.
isolates of Klebsiella pneumoniae emerging in intensive care unit
Probenecid-procaine penicillin G combination and tetracycline
patients: first report of a multiclonal cluster. J Antimicrob Chemother
hydrochloride in the treatment of “penicillin-resistant” gonorrhea in
2007; 59: 786–90.
men. JAMA 1967; 202: 461–73.
28 Baadani AM, Thawadi SI, El-Khizzi NA, Omrani AS. Prevalence of
6 Rasnake MS, Conger NG, McAllister K, Holmes KK, Tramont EC.
colistin and tigecycline resistance in Acinetobacter baumannii clinical
History of U.S. military contributions to the study of sexually
isolates from 2 hospitals in Riyadh Region over a 2-year period.
transmitted diseases. Mil Med 2005; 170 (suppl): 61–65.
Saudi Med J 2013; 34: 248–53.
7 Nordmann P, Naas T, Poirel L. Global spread of Carbapenemase-
29 Napier BA, Burd EM, Satola SW, et al. Clinical use of colistin induces
producing Enterobacteriaceae. Emerg Infect Dis 2011; 17: 1791–98.
cross-resistance to host antimicrobials in Acinetobacter baumannii.
8 Walsh TR, Weeks J, Livermore DM, Toleman MA. Dissemination of MBio 2013; 4: e00021–13.
NDM-1 positive bacteria in the New Delhi environment and its
30 Klein EY, Sun L, Smith DL, Laxminarayan R. The changing
implications for human health: an environmental point prevalence
epidemiology of methicillin-resistant Staphylococcus aureus in the United
study. Lancet Infect Dis 2011; 11: 355–62.
States: a national observational study. Am J Epidemiol 2013; 177: 666–74.
9 Ruiz J, Pons MJ, Gomes C. Transferable mechanisms of quinolone
31 Anupurba S, Sen MR, Nath G, Sharma BM, Gulati AK,
resistance. Int J Antimicrob Agents 2012; 40: 196–203.
Mohapatra TM. Prevalence of methicillin resistant Staphylococcus
10 Holden MT, Hsu LY, Kurt K, et al. A genomic portrait of the aureus in a tertiary referral hospital in eastern Uttar Pradesh.
emergence, evolution, and global spread of a methicillin-resistant Indian J Med Microbiol 2003; 21: 49–51.
Staphylococcus aureus pandemic. Genome Res 2013; 23: 653–64.
32 Zafar A, Stone M, Ibrahim S, et al. Prevalent genotypes of
11 Ammerlaan HS, Harbarth S, Buiting AG, et al. Secular trends in meticillin-resistant Staphylococcus aureus: report from Pakistan.
nosocomial bloodstream infections: antibiotic-resistant bacteria J Med Microbiol 2011; 60: 56–62.
increase the total burden of infection. Clin Infect Dis 2013;
33 Malik N, Butt T, Arfan-ul-Bari. Frequency and antimicrobial
56: 798–805.
susceptibility pattern of methicillin resistant Staphylococcus aureus.
12 Allegranzi B, Bagheri Nejad S, et al. Burden of endemic health-care- J Coll Physicians Surg Pak 2009; 19: 287–90.
associated infection in developing countries: systematic review and
34 ECDC/EMEA. Technical Report. The bacterial challenge: time to
meta-analysis. Lancet 2011; 377: 228–41.
react. Sept 2009. http://www.ecdc.europa.eu/en/publications/
13 Zaidi AK, Huskins WC, Thaver D, Bhutta ZA, Abbas Z, Publications/0909_TER_The_Bacterial_Challenge_Time_to_React.
Goldmann DA. Hospital-acquired neonatal infections in developing pdf (accessed Oct 31, 2013).
countries. Lancet 2005; 365: 1175–88.
35 Klevens RM, Morrison MA, Nadle J, et al, for the Active Bacterial Core
14 Waters D, Jawad I, Ahmad A, et al. Aetiology of community-acquired surveillance (ABCs) MRSA Investigators. Invasive methicillin-resistant
neonatal sepsis in low and middle income countries. J Glob Health Staphylococcus aureus infections in the United States. JAMA 2007;
2011; 1: 154–70. 298: 1763–71.

1090 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

36 US Centers for Disease Control and Prevention. Antibiotic resistance 57 Filippini M, Masiero G, Moschetti K. Socioeconomic determinants of
threats in the United States, 2013. April 2013. http://www.cdc.gov/ regional differences in outpatient antibiotic consumption: evidence
drugresistance/threat-report-2013/pdf/ar-threats-2013-508.pdf from Switzerland. Health Policy 2006; 78: 77–92.
(accessed Oct 31, 2013). 58 Glass SK, Pearl DL, McEwen SA, Finley R. Canadian province-level
37 Lambert ML, Suetens C, Savey A, et al. Clinical outcomes of risk factor analysis of macrolide consumption patterns (2000–2006).
health-care-associated infections and antimicrobial resistance in J Antimicrob Chemother 2010; 65: 148–55.
patients admitted to European intensive-care units: a cohort study. 59 Matuz M, Benko R, Doro P, et al. Regional variations in community
Lancet Infect Dis 2011; 11: 30–38. consumption of antibiotics in Hungary, 1996–2003.
38 Wolkewitz M, Frank U, Philips G, Schumacher M, Davey P, for the Br J Clin Pharmacol 2006; 61: 96–100.
BURDEN Study Group. Mortality associated with in-hospital 60 ECDC. Summary of the latest data on antibiotic consumption in the
bacteraemia caused by Staphylococcus aureus: a multistate analysis European Union. Nov 2012. http://ecdc.europa.eu/en/eaad/
with follow-up beyond hospital discharge. J Antimicrob Chemother Documents/ESAC-Net-summary-antibiotic-consumption.pdf
2011; 66: 381–86. (accessed Oct 31, 2013).
39 de Kraker ME, Davey PG, Grundmann H, for the BURDEN study 61 Morgan DJ, Okeke IN, Laxminarayan R, Perencevich EN,
group. Mortality and hospital stay associated with resistant Weisenberg S. Non-prescription antimicrobial use worldwide:
Staphylococcus aureus and Escherichia coli bacteremia: estimating the a systematic review. Lancet Infect Dis 2011; 11: 692–701.
burden of antibiotic resistance in Europe. PLoS Med 2011; 62 Sweidan M, Zhang Y, Harvey K, Yang Y, Shen X, Yao K. Proceedings
8: e1001104. of the 2nd national workshop on rational use of antibiotics in China.
40 Kayange N, Kamugisha E, Mwizamholya DL, Jeremiah S, Mshana SE. Beijing: Beijing Children’s Hospital, 2005.
Predictors of positive blood culture and deaths among neonates with 63 Dong H, Bogg L, Rehnberg C, Diwan V. Association between health
suspected neonatal sepsis in a tertiary hospital, Mwanza-Tanzania. insurance and antibiotics prescribing in four counties in rural China.
BMC Pediatr 2010; 10: 39. Health Policy 1999; 48: 29–45.
41 Office of Technology Assessment. Impacts of antibiotic-resistant 64 Bennett D, Lauderdale T-L, Hung C-L. Competing doctors, antibiotic
bacteria: a report to the US Congress. Washington DC: Government use, and antibiotic resistance in Taiwan. Sept 16, 2008. http://chess.
Printing Office, 1995. uchicago.edu/events/hew/fall08/bennett.pdf (accessed Nov 1, 2013).
42 Reed S, Sullivan S, Laxminarayan R. Socioeconomic issues related to 65 Apisarnthanarak A, Tunpornchai J, Tanawitt K, Mundy LM.
antibiotic use. In: Low DE, ed. Appropriate antibiotic use. London: Nonjudicious dispensing of antibiotics by drug stores in Pratumthani,
The Royal Society of Medicine Press, 2001. 41–46. Thailand. Infect Control Hosp Epidemiol 2008; 29: 572–75.
43 Holmberg SD, Solomon SL, Blake PA. Health and economic impacts 66 Wise R. Antimicrobial resistance: priorities for action.
of antimicrobial resistance. Rev Infect Dis 1987; 9: 1065–78. J Antimicrob Chemother 2002; 49: 585–86.
44 The Genesis Report. The real war on drugs: bacteria are winning. 67 Glad T, Bernhardsen P, Nielsen KM, et al. Bacterial diversity in faeces
New Jersey; December 1994. from polar bear (Ursus maritimus) in Arctic Svalbard. BMC Microbiol
45 Howard D, Rask K. The Impact of resistance on antibiotic demand in 2010; 10: 10.
patients with ear infections. In: Laxminarayan R, ed. Battling 68 Levy SB, FitzGerald GB, Macone AB. Spread of antibiotic-resistant
resistance to antibiotics and pesticides: an economic approach. plasmids from chicken to chicken and from chicken to man. Nature
Washington DC: RFF Press, 2002. 119–33. 1976; 260: 40–42.
46 O’Brien KL, Wolfson LJ, Watt JP, et al, for the Hib and Pneumococcal 69 Harrison EM, Paterson GK, Holden MT, et al. Whole genome
Global Burden of Disease Study Team. Burden of disease caused by sequencing identifies zoonotic transmission of MRSA isolates with
Streptococcus pneumoniae in children younger than 5 years: global the novel mecA homologue mecC. EMBO Mol Med 2013; 5: 509–15.
estimates. Lancet 2009; 374: 893–902.
70 Fleming A. Penicillin, its practical application. London: Butterworth
47 Gu B, Kelesidis T, Tsiodras S, Hindler J, Humphries RM. & Co, 1946.
The emerging problem of linezolid-resistant Staphylococcus.
71 Kunin CM. This is medical progress? Trends and consequences of
J Antimicrob Chemother 2013; 68: 4–11.
antibiotic use in the United States. In comment. JAMA 1974;
48 Patel SN, Memari N, Shahinas D, Toye B, Jamieson FB, Farrell DJ. 227: 1030–32.
Linezolid resistance in Enterococcus faecium isolated in Ontario,
72 Hranjec T, Rosenberger LH, Swenson B, et al. Aggressive versus
Canada. Diagn Microbiol Infect Dis 2013; published online Oct 2.
conservative initiation of antimicrobial treatment in critically ill
DOI:10.1016/j.diagmicrobio.2013.08.01
surgical patients with suspected intensive-care-unit-acquired
49 López-Salas P, Llaca-Díaz J, Morfin-Otero R, et al. Virulence and infection: a quasi-experimental, before and after observational cohort
antibiotic resistance of Enterococcus faecalis clinical isolates recovered study. Lancet Infect Dis 2012; 12: 774–80.
from three states of mexico. Detection of linezolid resistance.
73 Thriemer K, Katuala Y, Batoko B, et al. Antibiotic prescribing in DR
Arch Med Res 2013; 44: 422–28.
Congo: a knowledge, attitude and practice survey among medical
50 Laxminarayan R, Malani A, Howard D, Smith DL. Extending the doctors and students. PLoS ONE 2013; 8: e55495.
cure: policy responses to the growing threat of antibiotic resistance.
74 Global Antibiotic Resistance Partnership, Public Health Foundation
Washington DC: Resources for the Future, 2007.
of India. Situation analysis. Antibiotic use and resistance in India.
51 Smith R, Coast J. The true cost of antimicrobial resistance. BMJ 2013; March 2011. http://www.cddep.org/sites/cddep.org/files/publication_
346: 1493. files/India-report-web.pdf?issuusl=ignore (accessed Nov 4, 2013).
52 Goossens H, Ferech M, Vander Stichele R, Elseviers M, for the ESAC 75 Food and Drug Administration Department of Health and Human
Project Group. Outpatient antibiotic use in Europe and association Services. 2010 summary report on antimicrobials sold or distributed
with resistance: a cross-national database study. Lancet 2005; for use in food-producing animals. http://www.fda.gov/downloads/
365: 579–87. ForIndustry/UserFees/AnimalDrugUserFeeActADUFA/ucm277657.
53 Braykov NP, Eber MR, Klein EY, Morgan DJ, Laxminarayan R. pdf (accessed April 29, 2013).
Trends in resistance to carbapenems and third-generation 76 WHO. The evolving threat of antimicrobial resistance: option for
cephalosporins among clinical isolates of Klebsiella pneumoniae in action. 2012. http://www.who.int/patientsafety/implementation/amr/
the United States, 1999–2010. Infect Control Hosp Epidemiol 2013; publication/en/index.html (accessed May 24, 2013).
34: 259–68.
77 Stone SP, Fuller C, Savage J, et al. Evaluation of the national
54 Klein E, Smith DL, Laxminarayan R. Hospitalizations and deaths Cleanyourhands campaign to reduce Staphylococcus aureus
caused by methicillin-resistant Staphylococcus aureus, United States, bacteraemia and Clostridium difficile infection in hospitals in
1999–2005. Emerg Infect Dis 2007; 13: 1840–46. England and Wales by improved hand hygiene: four year, prospective,
55 Steinman MA, Landefeld CS, Gonzales R. Predictors of ecological, interrupted time series study. BMJ 2012; 344: e3005.
broad-spectrum antibiotic prescribing for acute respiratory tract 78 Schwaber MJ, Lev B, Israeli A, et al, for the Israel Carbapenem-
infections in adult primary care. JAMA 2003; 289: 719–25. Resistant Enterobacteriaceae Working Group. Containment of a
56 Nitzan O, Low M, Lavi I, Hammerman A, Klang S, Raz R. country-wide outbreak of carbapenem-resistant Klebsiella pneumoniae
Variability in outpatient antimicrobial consumption in Israel. in Israeli hospitals via a nationally implemented intervention.
Infection 2010; 38: 12–18. Clin Infect Dis 2011; 52: 848–55.

www.thelancet.com/infection Vol 13 December 2013 1091

 The Lancet Infectious Diseases Commission

79 Brinsley K, Srinivasan A, Sinkowitz-Cochran R, et al. Implementation 102 Wirtz VJ, Dreser A, Gonzales R. Trends in antibiotic utilization in
of the Campaign to Prevent Antimicrobial Resistance in Healthcare eight Latin American countries, 1997-2007. Rev Panam Salud Publica
Settings: 12 steps to prevent antimicrobial resistance among 2010; 27: 219–25.
hospitalized adults-experiences from 3 institutions. 103 Earnshaw S, Monnet DL, Duncan B, O’Toole J, Ekdahl K,
Am J Infect Control 2005; 33: 53–54. Goossens H, for the European Antibiotic Awareness Day Technical
80 Ghafur A, Mathai D, Muruganathan A, et al. The Chennai Advisory Committee, and the European Antibiotic Awareness Day
Declaration: a roadmap to tackle the challenge of antimicrobial Collaborative Group. European Antibiotic Awareness Day, 2008–the
resistance. Indian J Cancer 2013; 50: 71–73. first Europe-wide public information campaign on prudent
81 Gould IM, Wang GH, Wu J-J, et al. Global burden of multi drug antibiotic use: methods and survey of activities in participating
resistant organisms current antimicrobial resistance problems in the countries. Euro Surveill 2009; 14: 19280.
Asia-Pacific. MDRO Beijing Consensus Meeting Report. 104 Blozik E, Wildeisen IE, Fueglistaler P, von Overbeck J. Telemedicine
J Global Antimicrob Resistance (in press). can help to ensure that patients receive timely medical care.
82 Paterson DL. The role of antimicrobial management programs in J Telemed Telecare 2012; 18: 119–21.
optimizing antibiotic prescribing within hospitals. Clin Infect Dis 105 Minzi O, Manyilizu V. Application of basic pharmacology and
2006; 42 (suppl 2): 90–95. dispensing practice of antibiotics in accredited drug-dispensing
83 Lawes T, Edwards B, Lopez-Lozano JM, Gould I. Trends in outlets in Tanzania. Drug Healthc Patient Saf 2013; 5: 5–11.
Staphylococcus aureus bacteraemia and impacts of infection control 106 WHO. Promoting rational use of medicine: core components. WHO
practices including universal MRSA admission screening in a policy perspectives on medicine. Number 5. Sept 2002. http://apps.
hospital in Scotland, 2006–2010: retrospective cohort study and who.int/medicinedocs/en/d/Jh3011e/ (accessed Nov 1, 2013).
time-series intervention analysis. BMJ Open 2012; 2: 3. 107 Pulcini C, Gyssens IC. How to educate prescribers in antimicrobial
84 Vernaz N, Hill K, Leggeat S, et al. Temporal effects of antibiotic use stewardship practices. Virulence 2013; 4: 192–202.
and Clostridium difficile infections. J Antimicrob Chemother 2009; 108 Laxminarayan R, Klugman KP. Communicating trends in resistance
63: 1272–75. using a drug resistance index. BMJ Open 2011; 1: e000135.
85 Dellit TH, Owens RC, McGowan JE Jr, et al, for the Infectious 109 Huttner B, Goossens H, Verheij T, Harbarth S, for the CHAMP
Diseases Society of America, and the Society for Healthcare consortium. Characteristics and outcomes of public campaigns aimed
Epidemiology of America. Infectious Diseases Society of America and at improving the use of antibiotics in outpatients in high-income
the Society for Healthcare Epidemiology of America guidelines for countries. Lancet Infect Dis 2010; 10: 17–31.
developing an institutional program to enhance antimicrobial 110 Park S, Zhang Y, Lin S, Wang TH, Yang S. Advances in microfluidic
stewardship. Clin Infect Dis 2007; 44: 159–77. PCR for point-of-care infectious disease diagnostics. Biotechnol Adv
86 Nabyonga Orem J, Bataringaya Wavamunno J, Bakeera SK, Criel B. 2011; 29: 830–39.
Do guidelines influence the implementation of health programs? 111 Diekema DJ, Pfaller MA. Rapid detection of antibiotic-resistant
Uganda’s experience. Implement Sci 2012; 7: 98. organism carriage for infection prevention. Clin Infect Dis 2013;
87 Wertheim HFL, Chandna A, Vu PD, et al. Providing impetus, tools, 56: 1614–20.
and guidance to strengthen national capacity for antimicrobial 112 Galar A, Leiva J, Espinosa M, Guillén-Grima F, Hernáez S, Yuste JR.
stewardship in Viet Nam. PLoS Med 2013; 10: e1001429. Clinical and economic evaluation of the impact of rapid
88 Millar M. Constraining the use of antibiotics: applying Scanlon’s microbiological diagnostic testing. J Infect 2012; 65: 302–09.
contractualism. J Med Ethics 2012; 38: 465–69. 113 Kerremans JJ, Verboom P, Stijnen T, et al. Rapid identification and
89 Apisarnthanarak A, Mundy LM. Comparison of methods of antimicrobial susceptibility testing reduce antibiotic use and
measuring pharmacy sales of antibiotics without prescriptions in accelerate pathogen-directed antibiotic use. J Antimicrob Chemother
Pratumthani, Thailand. Infect Control Hosp Epidemiol 2009; 2008; 61: 428–35.
30: 1130–32. 114 Cantón R, Morosini MI. Emergence and spread of antibiotic
90 Radyowijati A, Haak H. Improving antibiotic use in low-income resistance following exposure to antibiotics. FEMS Microbiol Rev 2011;
countries: an overview of evidence on determinants. Soc Sci Med 35: 977–91.
2003; 57: 733–44. 115 Johnson AP, Davies J, Guy R, et al. Mandatory surveillance of
91 Thorgeirsson T, Kawachi I. Behavioral economics: merging methicillin-resistant Staphylococcus aureus (MRSA) bacteraemia in
psychology and economics for lifestyle interventions. Am J Prev Med England: the first 10 years. J Antimicrob Chemother 2012;
2013; 44: 185–89. 67: 802–09.
92 Sumpradit N, Chongtrakul P, Anuwong K, et al. Antibiotics Smart 116 Okeke I. Divining without seeds. The case for strengthening
Use: a workable model for promoting the rational use of medicines laboratory medicine in Africa. Ithaca: Cornell University Press, 2011.
in Thailand. Bull World Health Organ 2012; 90: 905–13. 117 European Centre for Disease Prevention and Control (ECDC).
93 McDonnell Norms Group. Antibiotic overuse: the influence of social European Antimicrobial Resistance Surveillance Network. http://
norms. J Am Coll Surg 2008; 207: 265–75. www.ecdc.europa.eu/en/activities/surveillance/EARS-Net/Pages/
94 MacDougall C, Polk RE. Antimicrobial stewardship programs in index.aspx (accessed Nov 1, 2013).
health care systems. Clin Microbiol Rev 2005; 18: 638–56. 118 WHO. Impacts of antimicrobial growth promoter termination in
95 Choi S, Reich M. Report cards and prescribing behavior: assessing Denmark. The WHO international review panel’s evaluation of the
the impacts of public disclosure on antibiotic prescribing rates in termination of the use of antimicrobial growth promoters in
South Korea. Third International Conference in Improving Use of Denmark. Foulum, Denmark, Nov 6–9, 2002. Geneva: World Health
Medicines; Antalya, Turkey; Nov14–18, 2011. 13 AD. Organization, 2003.
96 Masiero G, Filippini M, Ferech M, Goossens H. Socioeconomic 119 Food and Agriculture Organization (FAO), WHO, and World
determinants of outpatient antibiotic use in Europe. Organisation for Animal Health (OIE). Report of the FAO/WHO/OIE
Int J Public Health 2010; 55: 469–78. expert meeting on critically important antimicrobials. Nov 26–30,
97 Mintzes B. Pharmaceutical promotion and prevention of antibiotic 2007, Rome: FAO; Geneva: World Health Organization, 2008.
resistance. Mastering anti-infective therapies. Paris: 2005. 120 Bouki C, Venieri D, Diamadopoulos E. Detection and fate of antibiotic
98 Hawkings NJ, Butler CC, Wood F. Antibiotics in the community: resistant bacteria in wastewater treatment plants: a review.
a typology of user behaviours. Patient Educ Couns 2008; 73: 146–52. Ecotoxicol Environ Saf 2013; 91: 1–9.
99 Grigoryan L, Burgerhof JG, Degener JE, et al, for the Self-Medication 121 Shi P, Jia S, Zhang XX, Zhang T, Cheng S, Li A. Metagenomic
with Antibiotics and Resistance (SAR) Consortium. Determinants of insights into chlorination effects on microbial antibiotic resistance in
self-medication with antibiotics in Europe: the impact of beliefs, drinking water. Water Res 2013; 47: 111–20.
country wealth and the healthcare system. J Antimicrob Chemother 122 Davey P, Brown E, Charani E, et al. Interventions to improve
2008; 61: 1172–79. antibiotic prescribing practices for hospital inpatients.
100 Planta MB. The role of poverty in antimicrobial resistance. Cochrane Database Syst Rev 2013; 4: CD003543.
J Am Board Fam Med 2007; 20: 533–39. 123 DeNap JC, Hergenrother PJ. Bacterial death comes full circle:
101 Zarb P, Goossens H. Human use of antimicrobial agents. targeting plasmid replication in drug-resistant bacteria.
Rev Sci Tech 2012; 31: 121–33. Org Biomol Chem 2005; 3: 959–66.

1092 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

124 Kurinčič M, Klančnik A, Smole Možina S. Epigallocatechin gallate as 146 Nicks BA, Manthey DE, Fitch MT. The Centers for Medicare and
a modulator of Campylobacter resistance to macrolide antibiotics. Medicaid Services (CMS) community-acquired pneumonia core
Int J Antimicrob Agents 2012; 40: 467–71. measures lead to unnecessary antibiotic administration by emergency
125 Potera C. Phage renaissance: new hope against antibiotic resistance. physicians. Acad Emerg Med 2009; 16: 184–87.
Environ Health Perspect 2013; 121: a48–53. 147 Stokstad ELR, Jukes TH. Further observations on the animal protein
126 Allerberger F, Gareis R, Jindrák V, Struelens MJ. Antibiotic factor. Proc Soc Biol Exp Med 1949; 73: 523–28.
stewardship implementation in the EU: the way forward. 148 Swedish Ministry of Agriculture. Antimicrobial feed additives. Report
Expert Rev Anti Infect Ther 2009; 7: 1175–83. from the commission on antimicrobial feed additives. Stockholm,
127 Bodansky D. Whither the Kyoto Protocol? Durban and beyond. 1997. 139.
Aug 2011. http://belfercenter.ksg.harvard.edu/publication/21314/ 149 Robinson P. Controlled trial of aureomycin in premature twins and
whither_the_kyoto_protocol_durban_and_beyond.html (accessed triplets. Lancet 1952; 259: 52.
Oct 7, 2013) 150 Swann MM. Joint Committee on the use of antibiotics in animal
128 Cantón R, Akóva M, Carmeli Y, et al, for the European Network on husbandry and veterinary medicine. London: Her Majesty’s
Carbapenemases. Rapid evolution and spread of carbapenemases Stationery Office, 1969.
among Enterobacteriaceae in Europe. Clin Microbiol Infect 2012; 151 Grave K, Jensen VF, Odensvik K, Wierup M, Bangen M. Usage of
18: 413–31. veterinary therapeutic antimicrobials in Denmark, Norway and
129 Butler CC, Hood K, Verheij T, et al. Variation in antibiotic prescribing Sweden following termination of antimicrobial growth promoter use.
and its impact on recovery in patients with acute cough in primary Prev Vet Med 2006; 75: 123–32.
care: prospective study in 13 countries. BMJ 2009; 338: 2242. 152 Marshall BM, Levy SB. Food animals and antimicrobials: impacts on
130 Little P, Stuart B, Moore M, et al, for the GRACE consortium. human health. Clin Microbiol Rev 2011; 24: 718–33.
Amoxicillin for acute lower-respiratory-tract infection in primary care 153 Witte W. Selective pressure by antibiotic use in livestock.
when pneumonia is not suspected: a 12-country, randomised, Int J Antimicrob Agents 2000; 16 (suppl 1): 19–24.
placebo-controlled trial. Lancet Infect Dis 2013; 13: 123–29. 154 Aarestrup FM, Jensen VF, Emborg HD, Jacobsen E, Wegener HC.
131 Kang CI, Kim SH, Kim HB, et al. Pseudomonas aeruginosa bacteremia: Changes in the use of antimicrobials and the effects on productivity
risk factors for mortality and influence of delayed receipt of effective of swine farms in Denmark. Am J Vet Res 2010; 71: 726–33.
antimicrobial therapy on clinical outcome. Clin Infect Dis 2003; 155 Pagel SW, Gautier P. Use of antimicrobial agents in livestock.
37: 745–51. Rev Sci Tech 2012; 31: 145–88.
132 Ishengoma DR, Rwegoshora RT, Mdira KY, et al. Health laboratories 156 European Medicines Agency. Sales of veterinary antimicrobial agents
in the Tanga region of Tanzania: the quality of diagnostic services for in 26 EU/EEA countries in 2011. Third ESVAC report. Oct 15, 2013.
malaria and other communicable diseases. Ann Trop Med Parasitol http://www.ema.europa.eu/docs/en_GB/document_library/
2009; 103: 441–53. Report/2013/10/WC500152311.pdf (accessed Oct 16, 2013).
133 Kreft R, Costerton JM, Ehrlich GD. PCR is changing clinical 157 European Food Safety Authority (EFSA) and European Centre for
diagnostics. Microbe 2013; 8: 15–20. Disease Prevention and Control (ECDC). The European Union
134 Ritzi-Lehnert M. Development of chip-compatible sample preparation summary report on antimicrobial resistance in zoonotic and indicator
for diagnosis of infectious diseases. Expert Rev Mol Diagn 2012; bacteria from humans, animals and food in 2011. EFSA J 2013; 11: 3196.
12: 189–206. 158 Kruse H, Sørum H. Transfer of multiple drug resistance plasmids
135 Okeke IN, Peeling RW, Goossens H, et al. Diagnostics as essential between bacteria of diverse origins in natural microenvironments.
tools for containing antibacterial resistance. Drug Resist Updat 2011; Appl Environ Microbiol 1994; 60: 4015–21.
14: 95–106. 159 Wellington EM, Boxall AB, Cross P, et al. The role of the natural
136 Peeling RW, Smith PG, Bossuyt PMM. A guide for diagnostic environment in the emergence of antibiotic resistance in
evaluations. Nat Rev Microbiol 2008; 6: 2–6. gram-negative bacteria. Lancet Infect Dis 2013; 13: 155–65.
137 Loens K, Beck T, Ursi D, et al. Two quality control exercises involving 160 Catry B, Van Duijkeren E, Pomba MC, et al, for the Scientific
nucleic acid amplification methods for detection of Mycoplasma Advisory Group on Antimicrobials (SAGAM). Reflection paper on
pneumoniae and Chlamydophila pneumoniae and carried out 2 years MRSA in food-producing and companion animals: epidemiology
apart (in 2002 and 2004). J Clin Microbiol 2006; 44: 899–908. and control options for human and animal health. Epidemiol Infect
138 Loens K, van Loon AM, Coenjaerts F, et al, for the GRACE Study 2010; 138: 626–44.
Group. Performance of different mono- and multiplex nucleic acid 161 Holmberg SD, Osterholm MT, Senger KA, Cohen ML. Drug-resistant
amplification tests on a multipathogen external quality assessment Salmonella from animals fed antimicrobials. N Engl J Med 1984;
panel. J Clin Microbiol 2012; 50: 977–87. 311: 617–22.
139 World Economic Forum. The financial sustainability of health 162 Hald T, Lo Fo Wong DM, Aarestrup FM. The attribution of human
systems. A case for change. 2012. infections with antimicrobial resistant Salmonella bacteria in Denmark
140 Loens K, Van Heirstraeten L, Malhotra-Kumar S, Goossens H, to sources of animal origin. Foodborne Pathog Dis 2007; 4: 313–26.
Ieven M. Optimal sampling sites and methods for detection of 163 Nordstrom L, Liu CM, Price LB. Foodborne urinary tract infections:
pathogens possibly causing community-acquired lower respiratory a new paradigm for antimicrobial-resistant foodborne illness.
tract infections. J Clin Microbiol 2009; 47: 21–31. Front Microbiol 2013; 4: 29.
141 Gazin M, Paasch F, Goossens H, Malhotra-Kumar S, for the MOSAR 164 Chee-Sanford JC, Mackie RI, Koike S, et al. Fate and transport of
WP2 and SATURN WP1 Study Teams. Current trends in culture- antibiotic residues and antibiotic resistance genes following land
based and molecular detection of extended-spectrum-β-lactamase- application of manure waste. J Environ Qual 2009; 38: 1086–108.
harboring and carbapenem-resistant Enterobacteriaceae. 165 Chaslus-Dancla E, Lafont JP, Martel JL. Spread of resistance from
J Clin Microbiol 2012; 50: 1140–46. food animals to man: the French experience. Acta Vet Scand Suppl
142 Livermore DM, Andrews JM, Hawkey PM, et al. Are susceptibility 2000; 93 (suppl): 53–60.
tests enough, or should laboratories still seek ESBLs and 166 Lester CH, Frimodt-Møller N, Sørensen TL, Monnet DL, Hammerum
carbapenemases directly? J Antimicrob Chemother 2012; 67: 1569–77. AM. In vivo transfer of the vanA resistance gene from an
143 Falagas ME, Tansarli GS, Rafailidis PI, Kapaskelis A, Vardakas KZ. Enterococcus faecium isolate of animal origin to an E. faecium isolate
Impact of antibiotic MIC on infection outcome in patients with of human origin in the intestines of human volunteers.
susceptible Gram-negative bacteria: a systematic review and Antimicrob Agents Chemother 2006; 50: 596–99.
meta-analysis. Antimicrob Agents Chemother 2012; 56: 4214–22. 167 Seiffert SN, Hilty M, Perreten V, Endimiani A. Extended-spectrum
144 Little P, Stuart B, Francis N, et al. Effects of internet-based training on cephalosporin-resistant Gram-negative organisms in livestock: an
antibioitic prescribing rates for acute respiratory-tract infection: emerging problem for human health? Drug Resist Updat 2013;
a multinational, cluster, randomised, factorial, controlled trial. 16: 22–45.
Lancet 2013; 382: 175–82. 168 Leverstein-van Hall MA, Dierikx CM, Cohen Stuart J, et al, for the
145 Mitsuma SF, Mansour MK, Dekker JP, et al. Promising new assays National ESBL surveillance group. Dutch patients, retail chicken meat
and technologies for the diagnosis and management of infectious and poultry share the same ESBL genes, plasmids and strains.
diseases. Clin Infect Dis 2013; 56: 996–1002. Clin Microbiol Infect 2011; 17: 873–80.

www.thelancet.com/infection Vol 13 December 2013 1093

 The Lancet Infectious Diseases Commission

169 Börjesson S, Jernberg C, Brolund A, et al. Characterization of 193 Michel CE, Solomon AW, Magbanua JP, et al. Field evaluation of
plasmid-mediated AmpC-producing E. coli from Swedish broilers a rapid point-of-care assay for targeting antibiotic treatment for
and association with human clinical isolates. Clin Microbiol Infect trachoma control: a comparative study. Lancet 2006; 367: 1585–90.
2013; 19: 309–11. 194 Save the Children. Surviving the first day: state of the world’s
170 Fischer J, Rodríguez I, Schmoger S, et al. Salmonella enterica subsp. mothers. May 2013. http://www.savethechildren.org.uk/sites/default/
enterica producing VIM-1 carbapenemase isolated from livestock files/images/State_of_World_Mothers_2013.pdf (accessed Sept 24,
farms. J Antimicrob Chemother 2013; 68: 478–80. 2013).
171 Codex Alimentarius Commission. Guidelines for risk analysis of 195 Johansson EW, Carvajal L, Newby H, Young M, Wardlaw T.
foodborne antimicrobial resistance. http://www.foedevarestyrelsen. Pneumonia and diarrhoea: tackling the deadliest diseases for the
dk/SiteCollectionDocuments/25_PDF_word_filer%20til%20 world’s poorest children. June 2012. New York: UNICEF, 2012.
download/04kontor/resistens%20codex.pdf (accessed Nov 1, 2013). 196 Baqui AH, Black RE, El Arifeen S, et al. Zinc therapy for diarrhoea
172 Ashbolt NJ, Amézquita A, Backhaus T, et al. Human Health Risk increased the use of oral rehydration therapy and reduced the use of
Assessment (HHRA) for environmental development and transfer of antibiotics in Bangladeshi children. J Health Popul Nutr 2004;
antibiotic resistance. Environ Health Perspect 2013; 121: 993–1001. 22: 440–42.
173 WHO. Critically important antimicrobials for human medicine. 3rd 197 Rutschmann OT, Domino ME. Antibiotics for upper respiratory tract
revision 2011. http://www.who.int/foodborne_disease/resistance/cia/ infections in ambulatory practice in the United States, 1997–1999:
en (accessed Nov 1, 2013). does physician specialty matter? J Am Board Fam Pract 2004;
174 Smith DL, Harris AD, Johnson JA, Silbergeld EK, Morris JG Jr. 17: 196–200.
Animal antibiotic use has an early but important impact on the 198 Shapiro DJ, Hicks LA, Pavia AT, Hersh AL. Antibiotic
emergence of antibiotic resistance in human commensal bacteria. prescribing for adults in ambulatory care in the USA, 2007–09.
Proc Natl Acad Sci USA 2002; 99: 6434–39. J Antimicrob Chemother 2013; published online July 25.
175 Bowen M. Antimicrobial stewardship: Time for change. Equine Vet J DOI:10.1093/jac/dkt301.
2013; 45: 127–29. 199 Mainous AG 3rd, Everett CJ, Post RE, Diaz VA, Hueston WJ.
176 Cars O, Högberg LD, Murray M, et al. Meeting the challenge of Availability of antibiotics for purchase without a prescription on the
antibiotic resistance. BMJ 2008; 337: a1438. internet. Ann Fam Med 2009; 7: 431–35.
177 Carlet J, Jarlier V, Harbarth S, Voss A, Goossens H, Pittet D, for the 200 Blomberg B, Manji KP, Urassa WK, et al. Antimicrobial resistance
Participants of the 3rd World Healthcare-Associated Infections predicts death in Tanzanian children with bloodstream infections:
Forum. Ready for a world without antibiotics? The Pensières a prospective cohort study. BMC Infect Dis 2007; 7: 43.
Antibiotic Resistance Call to Action. Antimicrob Resist Infect Control 201 Siegel RM, Kiely M, Bien JP, et al. Treatment of otitis media with
2012; 1: 11. observation and a safety-net antibiotic prescription. Pediatrics 2003;
178 WHO. The rational use of drugs. Report of the Conference of 112: 527–31.
Experts, Nairobi, Nov 25-29, 1985. Geneva: World Health 202 Schiff GD, Wisniewski M, Bult J, Parada JP, Aggarwal H,
Organization, 1987. 67. Schwartz DN. Improving inpatient antibiotic prescribing: insights
179 Li Y, Xu J, Wang F, et al. Overprescribing in China, driven by financial from participation in a national collaborative. Jt Comm J Qual Improv
incentives, results in very high use of antibiotics, injections, and 2001; 27: 387–402.
corticosteroids. Health Aff (Millwood) 2012; 31: 1075–82. 203 Otrompke J. Public and private partnership helps to set the
180 Bigdeli M, Jacobs B, Tomson G, et al. Access to medicines from a standard of care for multi-drug resistant tuberculosis. Case studies
health system perspective. Health Policy Plan 2013; 28: 692–704. for global health. Oct 30, 2009. http://casestudiesforglobalhealth.
181 So AD, Gupta N, Brahmachari SK, et al. Towards new business org/post.cfm/public-and-private-partnership-helps-to-set-the-
models for R&D for novel antibiotics. Drug Resist Updat 2011; standard-of-care-for-multi-drug-resistant-tuberculosis-1 (accessed
14: 88–94. Nov 1, 2013).
182 So AD, Ruiz-Esparza Q, Gupta N, Cars O. 3Rs for innovating novel 204 Bisoffi Z, Sirima BS, Angheben A, et al. Rapid malaria diagnostic
antibiotics: sharing resources, risks, and rewards. BMJ 2012; tests vs. clinical management of malaria in rural Burkina Faso: safety
344: e1782. and effect on clinical decisions. A randomized trial.
Trop Med Int Health 2009; 14: 491–98.
183 Morones-Ramirez JR, Winkler JA, Spina CS, Collins JJ. Silver
enhances antibiotic activity against gram-negative bacteria. 205 Bastiaens GJH, Schaftenaar E, Ndaro A, Keuter M, Bousema T,
Sci Transl Med 2013; 5: 190ra81. Shekalaghe SA. Malaria diagnostic testing and treatment practices in
three different Plasmodium falciparum transmission settings in
184 Fischbach MA. Combination therapies for combating antimicrobial
Tanzania: before and after a government policy change. Malar J 2011;
resistance. Curr Opin Microbiol 2011; 14: 519–23.
10: 76.
185 Cottarel G, Wierzbowski J. Combination drugs, an emerging option
206 Batwala V, Magnussen P, Nuwaha F. Antibiotic use among patients
for antibacterial therapy. Trends Biotechnol 2007; 25: 547–55.
with febrile illness in a low malaria endemicity setting in Uganda.
186 Mishra RPN, Oviedo-Orta E, Prachi P, Rappuoli R, Bagnoli F. Malar J 2011; 10: 377.
Vaccines and antibiotic resistance. Curr Opin Microbiol 2012;
207 Nanyonjo A, Nakirunda M, Makumbi F, Tomson G, Källander K, for
15: 596–602.
the inSCALE Study Group. Community acceptability and adoption of
187 Goodman C, Gould C, Karnes E, et al. The value of diagnostics integrated community case management in Uganda.
innovation, adoption and diffusion into health care. The Lewin Group Am J Trop Med Hyg 2012; 87 (suppl 5): 97–104.
Reports. July 2005. http://www.lewin.com/~/media/Lewin/Site_
208 Mukanga D, Tiono AB, Anyorigiya T, et al. Integrated community
Sections/Publications/ValueofDiagnostics.pdf (accessed Nov 1, 2013).
case management of fever in children under five using rapid
188 WHO. Increasing access to diagnostics through technology transfer diagnostic tests and respiratory rate counting: a multi-country
and local production. Geneva: World Health Organization, 2011. cluster randomized trial. Am J Trop Med Hyg 2012;
189 Boehme CC, Nabeta P, Hillemann D, et al. Rapid molecular detection 87 (suppl 5): 21–29.
of tuberculosis and rifampin resistance. N Engl J Med 2010; 209 Hamer DH, Brooks ET, Semrau K, et al. Quality and safety of
363: 1005–15. integrated community case management of malaria using rapid
190 Lim YW, Steinhoff M, Girosi F, et al. Reducing the global burden of diagnostic tests and pneumonia by community health workers.
acute lower respiratory infections in children: the contribution of new Pathog Glob Health 2012; 106: 32–39.
diagnostics. Nature 2006; 444 (suppl 1): 9–18. 210 Waters HR, Rehwinkel JA, Burnham G. Economic evaluation of
191 Rinaldi A. Drug and a syphilis test offer hope of yaws eradication. Mectizan distribution. Trop Med Int Health 2004; 9: 16–25.
SciDevNet. April 24, 2013. http://www.scidev.net/global/disease/ 211 Sturchio JL. The case of ivermectin: lessons and implications for
news/drug-and-a-syphilis-test-offer-hope-of-yaws-eradication.html improving access to care and treatment in developing countries.
(accessed Nov 1, 2013). Community Eye Health 2001; 14: 22–23.
192 Mitjà O, Hays R, Ipai A, et al. Single-dose azithromycin versus 212 Trehan I, Goldbach HS, LaGrone LN, et al. Antibiotics as part of the
benzathine benzylpenicillin for treatment of yaws in children in management of severe acute malnutrition. N Engl J Med
Papua New Guinea: an open-label, non-inferiority, randomised trial. 2013; 368: 425–35.
Lancet 2012; 379: 342–47.

1094 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

213 Bole K. UCSF teams tackle childhood mortality and river blindness: 234 Kruse AY, Thieu Chuong DH, Phuong CN. Neonatal bloodstream
$16 million from Bill & Melinda Gates Foundation supports infections in a pediatric hospital in Vietnam: a cohort study.
international projects. UCSF News. Jan 10, 2013. http://www.ucsf. J Trop Pediatr 2013; published online July 18. DOI:10.1093/tropej/
edu/news/2013/01/13394/ucsf-teams-tackle-childhood-mortality-and- fmt056.
river-blindness (accessed Nov 1, 2013). 235 Elham AM. EL-Feky, Zeinab A, Rahman Y, Mansi YA. Retrospective
214 Rinaldi A. Drug and a syphilis test offer hope of yaws eradication. analysis of neonatal bacteremia and antimicrobial resistance pattern
April 24, 2013. http://www.scidev.net/global/disease/news/drug-and- in neonatal intensive care unit. Res J Med Med Sci 2011; 6: 62–68.
a-syphilis-test-offer-hope-of-yaws-eradication.html (accessed Nov 1, 236 Zaidi AK, Thaver D, Ali SA, Khan TA. Pathogens associated with
2013). sepsis in newborns and young infants in developing countries.
215 Khadem TM, Dodds Ashley E, Wrobel MJ, Brown J. Antimicrobial Pediatr Infect Dis J 2009; 28 (suppl 1): 10–18.
stewardship: a matter of process or outcome? Pharmacotherapy 2012; 237 West BA, Peterside O. Sensitivity pattern among bacterial isolates in
32: 688–706. neonatal septicaemia in port Harcourt. Ann Clin Microbiol Antimicrob
216 The Center for Disease Dynamics, Economics and Policy. Resistance 2012; 11: 7.
map. Antibiotic use and resistance: what’s new? http://www.cddep. 238 Lubell YAE, Ashley EA, Turner C, Turner P, White NJ. Susceptibility
org/ResistanceMap/key-findings#.UksSsmSG0Vk (accessed Oct 1, of community-acquired pathogens to antibiotics in Africa and Asia in
2013). neonates--an alarmingly short review. Trop Med Int Health 2011;
217 Institute for Healthcare Improvement. Science of improvement: 16: 145–51.
how to improve. April 24, 2011. http://www.ihi.org/knowledge/ 239 Lynch JP 3rd, Zhanel GG. Streptococcus pneumoniae: does
Pages/HowtoImprove/ScienceofImprovementHowtoImprove.aspx antimicrobial resistance matter? Semin Respir Crit Care Med 2009;
(accessed Nov 1, 2013). 30: 210–38.
218 Kilo CM. A framework for collaborative improvement: lessons from 240 Zafar A, Hussain Z, Lomama E, Sibille S, Irfan S, Khan E. Antibiotic
the Institute for Healthcare Improvement’s Breakthrough Series. susceptibility of pathogens isolated from patients with community-
Qual Manag Health Care 1998; 6: 1–13. acquired respiratory tract infections in Pakistan–the active study.
219 California Department of Public Health.The California antimicrobial Journal of Ayub Medical College. J Ayub Med Coll Abbottabad 2008;
stewardship program initiative. http://www.cdph.ca.gov/programs/ 20: 7–9.
hai/Pages/AntimicrobialStewardshipProgramInitiative.aspx 241 Borg MA, Tiemersma E, Scicluna E, et al, for the ARMed Project
(accessed Nov 1, 2013). members and collaborators. Prevalence of penicillin and
220 WHO. The evolving threat of antimicrobial resistance: options for erythromycin resistance among invasive Streptococcus pneumoniae
action. Geneva: World Health Organization, 2012. isolates reported by laboratories in the southern and eastern
221 Nugent R, Back E, Beith A. The race against drug resistance. Mediterranean region. Clin Microbiol Infect 2009; 15: 232–37.
Washington DC: Center for Global Development Washington, 2010. 242 Benouda A, Ben Redjeb S, Hammami A, Sibille S, Tazir M,
222 O’Brien TF. Emergence, spread, and environmental effect of Ramdani-Bouguessa N. Antimicrobial resistance of respiratory
antimicrobial resistance: how use of an antimicrobial anywhere can pathogens in North African countries. J Chemother 2009; 21: 627–32.
increase resistance to any antimicrobial anywhere else. Clin Infect Dis 243 Tali-Maamar H, Laliam R, Bentchouala C, et al. Serotyping and
2002; 34 (suppl 3): 78–84. antibiotic susceptibility of Streptococcus pneumoniae strains isolated in
223 Liu L, Johnson HL, Cousens S, et al, for the Child Health Epidemiology Algeria from 2001 to 2010. Med Mal Infect 2012; 42: 59–65.
Reference Group of WHO and UNICEF. Global, regional, and national 244 Agudelo CI, Castañeda E, Corso A, et al, for the Grupo Sireva II.
causes of child mortality: an updated systematic analysis for 2010 with Resistance to non-beta-lactam antibiotics in the clinical isolates of
time trends since 2000. Lancet 2012; 379: 2151–61. Streptococcus pneumoniae of children in Latin America. SIREVA II,
224 Okeke IN, Klugman KP, Bhutta ZA, et al. Antimicrobial resistance 2000-2005. Rev Panam Salud Publica 2009; 25: 305–13 (in Spanish).
in developing countries. Part II: strategies for containment. 245 Arifeen SE, Saha SK, Rahman S, et al. Invasive pneumococcal disease
Lancet Infect Dis 2005; 5: 568–80. among children in rural Bangladesh: results from a population-based
225 Monjur F, Rizwan F, Asaduzzaman M, et al. Antibiotic sensitivity surveillance. Clin Infect Dis 2009; 48 (suppl 2): 103–13.
pattern of causative organisms of neonatal septicemia in an urban 246 Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality
hospital of Bangladesh. Indian J Med Sci 2010; 64: 265–71. from 235 causes of death for 20 age groups in 1990 and 2010: a
226 Nepal HP, Acharya A, Gautam R, Shrestha S, Paudel R. systematic analysis for the Global Burden of Disease Study 2010.
Bacteriological profile of neonatal septicemia cases and the Lancet 2012; 380: 2095–128.
antimicrobial resistance pattern in a tertiary care hospital of central 247 Abera B, Bezabih B, Dessie A. Antimicrobial suceptibility of V.
Nepal. Internat J Biomed Research 2013; 4: 26–31. cholerae in north west, Ethiopia. Ethiop Med J 2010; 48: 23–28.
227 Shitaye D, Asrat D, Woldeamanuel Y, Worku B. Risk factors and 248 Balaji K, Okonjo PA, Thenmozhi R, Karutha Pandian S. Virulence
etiology of neonatal sepsis in Tikur Anbessa University Hospital, and multidrug resistance patterns of Vibrio cholerae O1 isolates from
Ethiopia. Ethiop Med J 2010; 48: 11–21. diarrheal outbreaks of South India during 2006–2009.
228 Chiabi A, Djoupomb M, Mah E, et al. The clinical and bacteriogical Microb Drug Resist 2013; 19: 198–203.
spectrum of neonatal sepsis in a tertiary hospital in yaounde, 249 Karki R, Bhatta DR, Malla S, et al. Resistotypes of Vibrio cholerae 01
cameroon. Iran J Pediatr 2011; 21: 441–48. Ogawa Biotype El Tor in Kathmandu, Nepal. Nepal Med Coll J 2011;
229 Acquah SEK, Quaye L, Sagoe K, Ziem JB, Bromberger PI, 13: 84–87.
Amponsem AA. Susceptibility of bacterial etiological agents to 250 Marin MA, Thompson CC, Freitas FS, et al. Cholera outbreaks in
commonly-used antimicrobial agents in children with sepsis at the Nigeria are associated with multidrug resistant atypical El Tor and
Tamale Teaching Hospital. BMC Infect Dis 2013; 13: 89. non-O1/non-O139 Vibrio cholerae. PLoS Negl Trop Dis 2013; 7: e2049.
230 Viswanathan R, Singh AK, Basu S, Chatterjee S, Sardar S, Isaacs D. 251 Sang WK, Oundo V, Schnabel D. Prevalence and antibiotic resistance
Multi-drug resistant gram negative bacilli causing early neonatal of bacterial pathogens isolated from childhood diarrhoea in four
sepsis in India. Archives of Dis in Childhood-Fetal and Neonatal Ed provinces of Kenya. J Infect Dev Ctries 2012; 6: 572–78.
2012; 97: 182–87. 252 Kiiru JN, Saidi SM, Goddeeris BM, Wamae NC, Butaye P,
231 Muhammad Z, Ahmed A, Hayat U, Wazir MS, Rafiyatullah WH. Kariuki SM. Molecular characterisation of Vibrio cholerae O1 strains
Neonatal sepsis: causative bacteria and their resistance to carrying an SXT/R391-like element from cholera outbreaks in Kenya:
antibiotics. Journal of Ayub Medical College. 1994–2007. BMC Microbiol 2009; 9: 275.
J Ayub Med Coll Abbottabad 2010; 22: 33–36. 253 Mwansa JCL, Mwaba J, Lukwesa C, et al. Multiply antibiotic-resistant
232 Awoniyi DO, Udo SJ, Oguntibeju OO. An epidemiological survey Vibrio cholerae O1 biotype El Tor strains emerge during cholera
of neonatal sepsis in a hospital in Western Nigeria. outbreaks in Zambia. Epidemiol Infect 2007; 135: 847–53.
Afr J Microbiol Research 2009; 3: 385–89. 254 Rashed SM, Mannan SB, Johura FT, et al. Genetic characteristics of
233 Al-Shamahy HA, Sabrah AA, Al-Robasi AB, Naser SM. Types of drug-resistant Vibrio cholerae O1 causing endemic cholera in Dhaka,
Bacteria associated with Neonatal Sepsis in Al-Thawra University 2006-2011. J Med Microbiol 2012; 61: 1736–45.
Hospital, Sana’a, Yemen, and their Antimicrobial Profile. 255 Charles RC, Ryan ET. Cholera in the 21st century. Curr Opin Infect Dis
Sultan Qaboos Univ Med J 2012; 12: 48–54. 2011; 24: 472–77.

www.thelancet.com/infection Vol 13 December 2013 1095

 The Lancet Infectious Diseases Commission

256 Borkakoty B, Biswas D, Devi U, Yadav K, Mahanta J. Emergence of 277 Vandenberg O, Nyarukweba DZ, Ndeba PM, et al. Microbiologic and
classical ctxB genotype 1 and tetracycline resistant strains of Vibrio clinical features of Salmonella species isolated from bacteremic
cholerae O1 El Tor in Assam, India. Trans R Soc Trop Med Hyg 2012; children in eastern Democratic Republic of Congo. Pediatr Infect Dis J
106: 382–86. 2010; 29: 504–10.
257 Shakoor S, Zaidi AK, Hasan R. Tropical bacterial gastrointestinal 278 Lunguya O, Lejon V, Phoba M-F, et al. Antimicrobial resistance in
infections. Infect Dis Clin North Am 2012; 26: 437–53. invasive non-typhoid Salmonella from the Democratic Republic of the
258 Gu B, Cao Y, Pan S, et al. Comparison of the prevalence and Congo: emergence of decreased fluoroquinolone susceptibility and
changing resistance to nalidixic acid and ciprofloxacin of Shigella extended-spectrum beta lactamases. PLoS Negl Trop Dis 2013; 7: e2103.
between Europe-America and Asia-Africa from 1998 to 2009. 279 Menezes GA, Khan MA, Harish BN, et al. Molecular characterization
Int J Antimicrob Agents 2012; 40: 9–17. of antimicrobial resistance in non-typhoidal salmonellae associated
259 Ghosh S, Pazhani GP, Chowdhury G, et al. Genetic characteristics with systemic manifestations from India. J Med Microbiol 2010;
and changing antimicrobial resistance among Shigella spp. isolated 59: 1477–83.
from hospitalized diarrhoeal patients in Kolkata, India. 280 Byarugaba DK. A view on antimicrobial resistance in developing
J Med Microbiol 2011; 60: 1460–66. countries and responsible risk factors. Int J Antimicrob Agents 2004;
260 Ahmed D, Hoque A, Elahi MSB, Endtz HP, Hossain MA. 24: 105–10.
Bacterial aetiology of diarrhoeal diseases and antimicrobial 281 Paruk F, Richards G, Scribante J, Bhagwanjee S, Mer M, Perrie H.
resistance in Dhaka, Bangladesh, 2005–2008. Epidemiol Infect 2012; Antibiotic prescription practices and their relationship to outcome in
140: 1678–84. South Africa: findings of the prevalence of infection in South African
261 Okeke IN, Laxminarayan R, Bhutta ZA, et al. Antimicrobial resistance intensive care units (PISA) study. S Afr Med J 2012; 102: 613–16.
in developing countries. Part I: recent trends and current status. 282 Datta S, Wattal C, Goel N, Oberoi JK, Raveendran R, Prasad KJ. A ten
Lancet Infect Dis 2005; 5: 481–93. year analysis of multi-drug resistant blood stream infections caused
262 Shea KM, and the American Academy of Pediatrics Committee on by Escherichia coli & Klebsiella pneumoniae in a tertiary care hospital.
Environmental Health, and the American Academy of Pediatrics Indian J Med Res 2012; 135: 907–12.
Committee on Infectious Diseases. Nontherapeutic use of 283 Rosenthal VD, Maki DG, Rodrigues C, et al, for the International
antimicrobial agents in animal agriculture: implications for Nosocomial Infection Control Consortium Investigators. Impact of
pediatrics. Pediatrics 2004; 114: 862–68. International Nosocomial Infection Control Consortium (INICC)
263 Srikantiah P, Girgis FY, Luby SP, et al. Population-based surveillance strategy on central line-associated bloodstream infection rates in the
of typhoid fever in Egypt. Am J Trop Med Hyg 2006; 74: 114–19. intensive care units of 15 developing countries.
264 Kumar S, Rizvi M, Berry N. Rising prevalence of enteric fever due to Infect Control Hosp Epidemiol 2010; 31: 1264–72.
multidrug-resistant Salmonella: an epidemiological study. 284 Laxminarayan R, Bhutta ZA, Duse A, et al. Drug resistance. In:
J Med Microbiol 2008; 57: 1247–50. Jamison DT, Breman JG, Measham AR, et al, eds. Disease control
265 Khan MI, Soofi SB, Ochiai RL, et al. Epidemiology, clinical priorities in developing countries, 2nd edn. New York: Oxford
presentation, and patterns of drug resistance of Salmonella Typhi in University Press, 2006. 1031–51.
Karachi, Pakistan. J Infect Dev Ctries 2012; 6: 704–14. 285 Byarugaba DK. Antimicrobial resistance and its containment in
266 Lunguya O, Lejon V, Phoba M-F, et al. Salmonella typhi in the developing countries. In: Antibiotic policies: theory and practice.
democratic republic of the congo: fluoroquinolone decreased Gould IM, van der Meer JWM, eds. New York: Springer, 2005. 617–47.
susceptibility on the rise. PLoS Negl Trop Dis 2012; 6: e1921. 286 Ashley EA, Lubell Y, White NJ, Turner P. Antimicrobial susceptibility
267 Kariuki S, Revathi G, Kiiru J, et al. Typhoid in Kenya is associated of bacterial isolates from community acquired infections in
with a dominant multidrug-resistant Salmonella enterica serovar Typhi Sub-Saharan Africa and Asian low and middle income countries.
haplotype that is also widespread in Southeast Asia. J Clin Microbiol Trop Med Int Health 2011; 16: 1167–79.
2010; 48: 2171–76. 287 Downie L, Armiento R, Subhi R, Kelly J, Clifford V, Duke T.
268 Mills-Robertson F, Addy ME, Mensah P, Crupper SS. Molecular Community-acquired neonatal and infant sepsis in developing
characterization of antibiotic resistance in clinical Salmonella typhi countries: efficacy of WHO’s currently recommended
isolated in Ghana. FEMS Microbiol Lett 2002; 215: 249–53. antibiotics-systematic review and meta-analysis. Arch Dis Child
2013; 98: 146–54.
269 Molyneux EM, Walsh AL, Malenga G, Rogerson S, Molyneux ME.
Salmonella meningitis in children in Blantyre, Malawi, 1996–1999. 288 Nugent R. Food and agriculture policy: issues related to prevention of
Ann Trop Paediatr 2000; 20: 41–44. noncommunicable diseases. Food Nutr Bull 2004; 25: 200–07.
270 Tabu C, Breiman RF, Ochieng B, et al. Differing burden and 289 Nugent R, Back E, Beith A, Center for Global Development’s Drug
epidemiology of non-Typhi Salmonella bacteremia in rural and urban Resistance Working Group. The race against drug resistance. June 14,
Kenya, 2006–2009. PLoS One 2012; 7: e31237. 2010. http://www.cgdev.org/publication/race-against-drug-resistance
(accessed Nov 1, 2013).
271 Feasey NA, Dougan G, Kingsley RA, Heyderman RS, Gordon MA.
Invasive non-typhoidal salmonella disease: an emerging and 290 Nugent R, Okeke IN. When medicines fail: recommendations for
neglected tropical disease in Africa. Lancet 2012; 379: 2489–99. curbing antibiotic resistance. J Infect Dev Ctries 2010; 4: 355–56.
272 Kariuki S, Revathi G, Kariuki N, Kiiru J, Mwituria J, Hart CA. 291 Transparency International. Global corruption report 2006: corruption
Characterisation of community acquired non-typhoidal Salmonella and health. London: Pluto Press, 2006.
from bacteraemia and diarrhoeal infections in children admitted to 292 Raka L. Prevention and Control of Hospital-Related Infections in
hospital in Nairobi, Kenya. BMC Microbiol 2006; 6: 101. Low and Middle Income Countries. Open Infect Dis J 2010; 4: 125–31.
273 Gordon MA, Graham SM, Walsh AL, et al. Epidemics of invasive 293 Goossens H. The Chennai declaration on antimicrobial resistance in
Salmonella enterica serovar enteritidis and S. enterica Serovar India. Lancet Infect Dis 2013; 13: 105–06.
typhimurium infection associated with multidrug resistance among 294 Leung E, Weil DE, Raviglione M, Nakatani H, for the World Health
adults and children in Malawi. Clin Infect Dis 2008; 46: 963–69. Organization World Health Day Antimicrobial Resistance Technical
274 Kingsley RA, Msefula CL, Thomson NR, et al. Epidemic multiple Working Group. The WHO policy package to combat antimicrobial
drug resistant Salmonella Typhimurium causing invasive disease in resistance. Bull World Health Organ 2011; 89: 390–92.
sub-Saharan Africa have a distinct genotype. Genome Res 2009; 295 Nishtar S. Choked pipes: reforming Pakistan’s mixed health system.
19: 2279–87. Oxford: University Press Oxford, 2010.
275 Bachou H, Tylleskär T, Kaddu-Mulindwa DH, Tumwine JK. 296 Ghana Statistical Service. Public expenditure tracking surveys. 2007.
Bacteraemia among severely malnourished children infected and http://www.statsghana.gov.gh/nada/index.php/catalog/57 (accessed
uninfected with the human immunodeficiency virus-1 in Kampala, Aug 13, 2013).
Uganda. BMC Infect Dis 2006; 6: 160. 297 Lewis M, Pettersson G, The World Bank Development Economics
276 Ikumapayi UN, Antonio M, Sonne-Hansen J, et al. Molecular Department and Human Development Department. Governance in
epidemiology of community-acquired invasive non-typhoidal health care delivery: raising performance. Oct 2009. http://www-wds.
Salmonella among children aged 2 29 months in rural Gambia and worldbank.org/servlet/WDSContentServer/WDSP/IB/2009/10/13/
discovery of a new serovar, Salmonella enterica Dingiri. J Med Microbiol 000158349_20091013151915/Rendered/PDF/WPS5074.pdf (accessed
2007; 56: 1479–84. Oct 7, 2013)

1096 www.thelancet.com/infection Vol 13 December 2013

 The Lancet Infectious Diseases Commission

298 Koziol M, Tolmie C. Using public expenditure tracking surveys to 321 Mossialos E, Morel CM, Edwards S, Berenson J,
monitor projects and small-scale programs: a guidebook. Aug 2010. Gemmill-Toyama M, Brogan D, World Health Organization 2010,
http://elibrary.worldbank.org/doi/abs/10.1596/978-0-8213-8519-7 on behalf of the European Observatory on Health Systems and
(accessed Nov 1, 2013). Policies. Policies and incentives for promoting innovation in
299 Freedominfo.org. Mexico. Freedom of information. http://www. antibiotic research. 2010. http://www.euro.who.int/__data/assets/
freedominfo.org/regions/latin-america/mexico/mexico2/ (accessed pdf_file/0011/120143/E94241.pdf (accessed Aug 17, 2013).
Aug 13, 2013). 322 McKenna M. Antibiotic resistance: Scandinavia gets it. Sept 8, 2010.
300 Ramkumar V. Our money, our responsibility: a citizens’ guide to http://www.superbugtheblog.com/2010/09/antibiotic-resistance-
monitoring government expenditures. Washington DC: scandinavia-gets.html (accessed Nov 1, 2013).
International Budget Project, 2008. 323 Ministry of Health and Social Affairs, Sweden. Innovative incentives
301 Transparency International. Corruption perceptions index 2012. for effective antibacterials. http://www.lakemedelsverket.se/upload/
http://www.transparency.org/ cpi2012/results (accessed Aug 12, nyheter/2009/Rapport%20fr%C3%A5n%20m%C3%B6tet.pdf
2013). (accessed Nov 1, 2013)
302 WHO. Developing health management information systems: a 324 European Commission. Communication from the Commission to
practical guide for developing countries. Geneva: The World Health the European Parliament and the Council. Action plan against the
Organization, 2004. rising threats from antimicrobial resistance. http://ec.europa.eu/dgs/
303 Fraser HS, Biondich P, Moodley D, Choi S, Mamlin BW, health_consumer/docs/communication_amr_2011_748_en.pdf
Szolovits P. Implementing electronic medical record systems in (accessed Nov 1, 2013).
developing countries. Inform Prim Care 2005; 13: 83–95. 325 Bergström R. The role of the pharmaceutical industry in meeting the
304 Dagan R. Impact of pneumococcal conjugate vaccine on infections public health threat of antibacterial resistance. Drug Resist Updat 2011;
caused by antibiotic-resistant Streptococcus pneumoniae. 14: 77–78.
Clin Microbiol Infect 2009; 15 (suppl 3): 16–20. 326 Haverfield S. New approaches to antibacterial drug development;
305 Ganguly NK, Arora NK, Chandy SJ, et al, for the Global Antibiotic request for comments. May 31, 2013. http://phrma.org/phrmapedia/
Resistance Partnership (GARP)–India Working Group. publications/testimony-public-comments/new-approaches-
Rationalizing antibiotic use to limit antibiotic resistance in India. antibacterial-drug-development (accessed Nov 1, 2013).
Indian J Med Res 2011; 134: 281–94. 327 Spellberg B. New antibiotic development: barriers and opportunities
306 Levin A, Kaddar M. Role of the private sector in the provision of in 2012. Alliance for the prudent use of antibiotics. Newsletter vol 30,
immunization services in low- and middle-income countries. issue 1. http://www.tufts.edu/med/apua/news/news-newsletter-vol-
Health Policy Plan 2011; 26 (suppl 1): 4–12. 30-no-1-2.shtml (accessed Nov 1, 2013).
307 Joice R, Lipsitch M. Targeting imperfect vaccines against 328 Infectious Diseases Society of America. The 10 x ’20 Initiative:
drug-resistance determinants: a strategy for countering the rise of pursuing a global commitment to develop 10 new antibacterial drugs
drug resistance. PLoS One 2013; 8: e68940. by 2020. Clin Infect Dis 2010; 50: 1081–83.
308 Gouws E, Bryce J, Habicht JP, et al. Improving antimicrobial use 329 Davies S, Chief Medical Officer, UK. Independent.co.uk, March 26,
among health workers in first-level facilities: results from the 2013.
multi-country evaluation of the Integrated Management of 330 Towse A, Sharma P. Incentives for R&D for new antimicrobial drugs.
Childhood Illness strategy. Bull World Health Organ 2004; Int J Econ Bus 2011; 18: 331–50.
82: 509–15. 331 Infectious Diseases Society of America. White paper:
309 Contessotto Spadetto C, Cámara Simón M, Avilés Inglés MJ, recommendations on the conduct of superiority and organism-
Ojeda Escuriet JM, Cascales Barceló I, Rodríguez Sánchez F. specific clinical trials of antibacterial agents for the treatment of
Rational use of antibiotics in pediatrics: impact of a rapid test for infections caused by drug-resistant bacterial pathogens. Clin Infect Dis
detection of beta-haemolytic group A streptococci in acute 2012; 55: 1031–46.
pharyngotonsillitis. An Esp Pediatr 2000; 52: 212–19 (in Spanish). 332 Goldfarb NM. Clinical operations: accelerating trials, allocating
310 Coates AR, Halls G, Hu Y. Novel classes of antibiotics or more of resources and measuring performance. J Clin Res Best Practices 2006;
the same? Br J Pharmacol 2011; 163: 184–94. 2: 12.
311 Powers JH. Antimicrobial drug development—the past, the 333 Coates A. Antibiotic discovery. Pan European Networks: Science and
present, and the future. Clin Microbiol Infect 2004; Technology 2012; 4: 1–2.
10 (suppl 4): 23–31. 334 Tralau-Stewart CJ, Wyatt CA, Kleyn DE, Ayad A. Drug discovery: new
312 Fenton C, Keating GM, Curran MP. Daptomycin. Drugs 2004; models for industry-academic partnerships. Drug Discov Today 2009;
64: 445–55. 14: 95–101.
313 Zappia G, Menendez P, Monache GD, Misiti D, Nevola L, Botta B. 335 Jenks S. New NIH center broadens scope of translational research.
The contribution of oxazolidinone frame to the biological activity of J Natl Cancer Inst 2012; 104: 728–31.
pharmaceutical drugs and natural products. Mini Rev Med Chem 336 Williams RJ, Walker I, Takle AK. Collaborative approaches to
2007; 7: 389–409. anticancer drug discovery and development: a Cancer Research UK
314 Boucher HW, Talbot GH, Bradley JS, et al. Bad bugs, no drugs: no perspective. Drug Discov Today 2012; 17: 185–87.
ESKAPE! An update from the Infectious Diseases Society of 337 Cohen J. Infectious disease. Approval of novel TB drug celebrated–
America. Clin Infect Dis 2009; 48: 1–12. with restraint. Science 2013; 339: 130.
315 Coates AR, Halls G. Antibiotics in phase II and III clinical trials. 338 Coates ARM. Antibiotic Resistance-them or us? Pan European
Handb Exp Pharmacol 2012; 211: 167–83. Networks. Science and Technology 2012. http://www.helperby.com/
316 Bax R and Griffin D. Introduction to antibiotic resistance. wp-content/uploads/2013/08/ST5-A-Coates-2023-CE-Atl.pdf (accessed
Handb Exp Pharmacol 2012; 211: 1–12. Nov 1, 2013).
317 Hernandez V, Crépin T, Palencia A, et al. Discovery of a novel class 339 Ballell L, Bates RH, Young RJ, et al. Fueling open-source drug
of boron-based antibacterials with activity against gram-negative discovery: 177 small-molecule leads against tuberculosis.
bacteria. Antimicrob Agents Chemother 2013; 57: 1394–403. ChemMedChem 2013; 8: 313–21.
318 Chen YH, Ko WC, Hsueh PR. Emerging resistance problems and 340 Cooper MA, Shlaes D. Fix the antibiotics pipeline. Nature 2011; 472: 32.
future perspectives in pharmacotherapy for complicated urinary 341 Fernebro J. Fighting bacterial infections-future treatment options.
tract infections. Expert Opin Pharmacother 2013; 14: 587–96. Drug Resist Updat 2011; 14: 125–39.
319 Kahlmeter G, Poulsen HO. Antimicrobial susceptibility of 342 Bartlett JG. Narrative review: the new epidemic of Clostridium
Escherichia coli from community-acquired urinary tract infections difficile-associated enteric disease. Ann Intern Med 2006; 145: 758–64.
in Europe: the ECO·SENS study revisited. Int J Antimicrob Agents 343 Rex JH, Eisenstein BI, Alder J, et al. A comprehensive regulatory
2012; 39: 45–51. framework to address the unmet need for new antibacterial
320 Widerström M, Wiström J, Ek E, Edebro H, Monsen T. Near treatments. Lancet Infect Dis 2013; 13: 269–75.
absence of methicillin-resistance and pronounced genetic diversity 344 Payne DJ, Gwynn MN, Holmes DJ, Pompliano DL. Drugs for bad
among Staphylococcus epidermidis isolated from healthy persons bugs: confronting the challenges of antibacterial discovery.
in northern Sweden. APMIS 2011; 119: 505–12. Nat Rev Drug Discov 2007; 6: 29–40.

www.thelancet.com/infection Vol 13 December 2013 1097

 The Lancet Infectious Diseases Commission

345 Bérdy J. Bioactive microbial metabolites. J Antibiot (Tokyo) 2005; 368 De Leon GP, Elowe NH, Koteva KP, Valvano MA, Wright GD. An in
58: 1–26. vitro screen of bacterial lipopolysaccharide biosynthetic enzymes
346 Bérdy J. Thoughts and facts about antibiotics: where we are now and identifies an inhibitor of ADP-heptose biosynthesis. Chem Biol 2006;
where we are heading. J Antibiot (Tokyo) 2012; 65: 385–95. 13: 437–41.
347 Donald RG, Skwish S, Forsyth RA, et al. A Staphylococcus aureus 369 Langsdorf EF, Malikzay A, Lamarr WA, et al. Screening for
fitness test platform for mechanism-based profiling of antibacterial antibacterial inhibitors of the UDP-3-O-(R-3-hydroxymyristoyl)-N-
compounds. Chem Biol 2009; 16: 826–36. acetylglucosamine deacetylase (LpxC) using a high-throughput mass
348 Eisenstein BI, Oleson FB Jr, Baltz RH. Daptomycin: from the spectrometry assay. J Biomol Screen 2010; 15: 52–61.
mountain to the clinic, with essential help from Francis Tally, MD. 370 Richter SG, Elli D, Kim HK, et al. Small molecule inhibitor of
Clin Infect Dis 2010; 50 (suppl 1): 10–15. lipoteichoic acid synthesis is an antibiotic for Gram-positive bacteria.
349 Brown ED. Is the GAIN Act a turning point in new antibiotic Proc Natl Acad Sci USA 2013; 110: 3531–36.
discovery? Can J Microbiol 2013; 59: 153–56. 371 Brodin P, Poquet Y, Levillain F, et al. High content phenotypic
350 Kalan L, Wright GD. Antibiotic adjuvants: multicomponent cell-based visual screen identifies Mycobacterium tuberculosis
anti-infective strategies. Expert Rev Mol Med 2011; 13: 5. acyltrehalose-containing glycolipids involved in phagosome
remodeling. PLoS Pathog 2010; 6: e1001100.
351 Drawz SM, Bonomo RA. Three decades of beta-lactamase inhibitors.
Clin Microbiol Rev 2010; 23: 160–201. 372 Reichert JM. Marketed therapeutic antibodies compendium.
MAbs 2012; 4: 413–15.
352 Hugonnet JE, Tremblay LW, Boshoff HI, Barry CE 3rd, Blanchard JS.
Meropenem-clavulanate is effective against extensively drug-resistant 373 Oleksiewicz MB, Nagy G, Nagy E. Anti-bacterial monoclonal
Mycobacterium tuberculosis. Science 2009; 323: 1215–18. antibodies: back to the future? Arch Biochem Biophys 2012;
526: 124–31.
353 Shlaes DM. New β-lactam-β-lactamase inhibitor combinations in
clinical development. Ann N Y Acad Sci 2013; 1277: 105–14. 374 Bebbington C, Yarranton G. Antibodies for the treatment of bacterial
infections: current experience and future prospects.
354 Shakya T, Stogios PJ, Waglechner N, et al. A small molecule
Curr Opin Biotechnol 2008; 19: 613–19.
discrimination map of the antibiotic resistance kinome. Chem Biol
2011; 18: 1591–601. 375 Hancock RE, Nijnik A, Philpott DJ. Modulating immunity as a
therapy for bacterial infections. Nat Rev Microbiol 2012; 10: 243–54.
355 Lomovskaya O, Bostian KA. Practical applications and feasibility of
efflux pump inhibitors in the clinic–a vision for applied use. 376 Hancock RE, Sahl HG. Antimicrobial and host-defense peptides as
Biochem Pharmacol 2006; 71: 910–18. new anti-infective therapeutic strategies. Nat Biotechnol 2006;
24: 1551–57.
356 Ejim L, Farha MA, Falconer SB, et al. Combinations of antibiotics and
nonantibiotic drugs enhance antimicrobial efficacy. Nat Chem Biol 377 Easton DM, Nijnik A, Mayer ML, Hancock RE. Potential of
2011; 7: 348–50. immunomodulatory host defense peptides as novel anti-infectives.
Trends Biotechnol 2009; 27: 582–90.
357 Spitzer M, Griffiths E, Blakely KM, et al. Cross-species discovery of
syncretic drug combinations that potentiate the antifungal 378 McFarland LV. Meta-analysis of probiotics for the prevention of
fluconazole. Mol Syst Biol 2011; 7: 499. antibiotic associated diarrhea and the treatment of Clostridium difficile
disease. Am J Gastroenterol 2006; 101: 812–22.
358 Farha MA, Leung A, Sewell EW, et al. Inhibition of WTA synthesis
blocks the cooperative action of PBPs and sensitizes MRSA to 379 Kassam Z, Lee CH, Yuan Y, Hunt RH. Fecal microbiota
β-lactams. ACS Chem Biol 2013; 8: 226–33. transplantation for Clostridium difficile infection: systematic review
and meta-analysis. Am J Gastroenterol 2013; 108: 500–08.
359 Rogers SA, Huigens RW 3rd, Cavanagh J, Melander C. Synergistic
effects between conventional antibiotics and 2-aminoimidazole- 380 Falup-Pecurariu O. Lessons learnt after the introduction of the seven
derived antibiofilm agents. Antimicrob Agents Chemother 2010; valent-pneumococcal conjugate vaccine toward broader spectrum
54: 2112–18. conjugate vaccines. Biom J 2012; 35: 450–56.
360 Wenderska IB, Chong M, McNulty J, Wright GD, Burrows LL. 381 McKenna M. Vaccine development: Man vs MRSA. Nature 2012;
Palmitoyl-DL-carnitine is a multitarget inhibitor of Pseudomonas 482: 23–25.
aeruginosa biofilm development. ChemBioChem 2011; 12: 2759–66. 382 Prabowo SA, Gröschel MI, Schmidt ED, et al. Targeting multidrug-
361 Escaich S. Antivirulence as a new antibacterial approach for resistant tuberculosis (MDR-TB) by therapeutic vaccines.
chemotherapy. Curr Opin Chem Biol 2008; 12: 400–08. Med Microbiol Immunol (Berl) 2013; 202: 95–104.
362 Mattmann ME, Blackwell HE. Small molecules that modulate 383 European Centres for Disease Prevention and Control and European
quorum sensing and control virulence in Pseudomonas aeruginosa. Medicines Agency. The bacterial challenge: time to react. Sept 2009.
J Org Chem 2010; 75: 6737–46. http://www.ecdc.europa.eu/en/publications/Publications/0909_TER_
The_Bacterial_Challenge_Time_to_React.pdf (accessed Nov 1, 2013).
363 Mansson M, Nielsen A, Kjærulff L, et al. Inhibition of virulence gene
expression in Staphylococcus aureus by novel depsipeptides from a 384 Wernli D, Haustein T, Conly J, Carmeli Y, Kickbusch I, Harbarth S.
marine photobacterium. Mar Drugs 2011; 9: 2537–52. A call for action: the application of The International Health
Regulations to the global threat of antimicrobial resistance. PLoS Med
364 Nielsen A, Nielsen KF, Frees D, Larsen TO, Ingmer H. Method for
2011; 8: e1001022.
screening compounds that influence virulence gene expression in
Staphylococcus aureus. Antimicrob Agents Chemother 2010; 54: 509–12. 385 Grundmann H, Klugman KP, Walsh T, et al. A framework for global
surveillance of antibiotic resistance. Drug Resist Updat 2011; 14: 79–87.
365 Hung DT, Shakhnovich EA, Pierson E, Mekalanos JJ. Small-molecule
inhibitor of Vibrio cholerae virulence and intestinal colonization. 386 World Bank. East Africa Public Health Laboratory Networking
Science 2005; 310: 670–74. Project. http://www-wds.worldbank.org/external/default/
WDSContentServer/WDSP/IB/2010/02/24/000020953_201002241502
366 Duncan MC, Linington RG, Auerbuch V. Chemical inhibitors of the
02/Rendered/PDF/531990PID0AFR0public0health0rev.pdf (accessed
type three secretion system: disarming bacterial pathogens.
Nov 1, 2013).
Antimicrob Agents Chemother 2012; 56: 5433–41.
367 Turgeon Z, Jørgensen R, Visschedyk D, et al. Newly discovered and
characterized antivirulence compounds inhibit bacterial mono-ADP-
ribosyltransferase toxins. Antimicrob Agents Chemother 2011;
55: 983–91.

1098 www.thelancet.com/infection Vol 13 December 2013