Antibiotic stewardship in the ICU

Considering the prevalence of multiresistant organisms in the ICU, and the tendency of intensivists to prescribe hardcore antibiotics, it is surprising that it has taken until 2022 for the college examiners to ask about it in the CICM Second Part papers. 

Question 2 from the first paper of 2022 was the first appearance of this topic. Curiously, the principles and practices were allocated only 20% of the mark, i.e. only two minutes (a short paragraph) was expected. The rest of the question was developed around the advantages and disadvantages of this practice, which is an interesting sideways glance from the examiners, suggesting that perhaps there is some hidden undercurrent of resentment towards this concept among the sort of people that routinely ask for high-dose vancopenem as their first choice of empiric therapy.

The LITFL page on antimicrobial stewardship is an excellent point-form summary to which everybody should refer before reading further, mainly as a sanity preservation mechanism. There is a good chance that all of the reader's needs will be satisfied by that quick revision item. Primary sources also include:

Definition of antibiotic stewardship

The term "stewardship", by its conventional definition (responsible planning, care and management) implies that antibiotic prescribing is something that needs to be carefully supervised. The CDC defines it as:

 "the effort to measure and improve how antibiotics are prescribed by clinicians and used by patients."


"coordinated interventions designed to improve and measure the appropriate use of antimicrobials"

In Australia, the Commission on Safety and Quality in Healthcare publish a regularly revised statement of standards (here's the 2020 version), and use a slightly different wording:

"an ongoing effort by a health service organisation to optimise antimicrobial use among patients"


"a systematic approach to optimising the use of antimicrobials, in order to preserve the effectiveness of these important medicines for treating infections."

In UK, the NHS boils it down to just the basics:

"the careful and responsible management of antimicrobial use"

Which is really just the dictionary definition of "stewardship".

What was the point of all this? Definitions are many, and this concept enjoys such widespread popularity that a junior ICU trainee is likely to already grasp it on some intuitive brainstem level, and they may never have had to think of a single phrase with which to describe this practice. The reader is invited to roll their own definition. As long as it includes key phrases like "systematic approach", "coordinated interventions" and "responsible management", it is likely to score marks.

Rationale for antibiotic stewardship

Borrowing from the key points raised by Kieran Hand (2013), the rationale for antibiotic stewardship can be broken down to two main arguments.

Argument from utilitarian ethics: 

  • Antibiotics are an extremely effective therapy for infectious diseases, which (prior to the availability of antibiotics) were a dominant cause of death globally.
  • The efficacy of antibiotics rests on their being able to overcome the mechanisms of antimicrobial resistance
  • Antimicrobial resistance develops as the consequence of exposure to antibiotics
  • Thus, antibiotic use is lifesaving for individual patients, but poses a public health threat when it leads to the emergence of multi-drug resistant organisms
  • The use of antibiotics for individual patients must therefore be carefully balanced against the risk to society

Argument from economics:

  • Antibiotic resistance can be overcome by the development of new antibiotics
  • However, drug discovery and development are expensive, and specifically the development of new antibiotics is unprofitable (when compared to the development of drugs intended to be used chronically)
  • Existing (older) antibiotic classes are often cheaper to manufacture than the newer classes 
  • To maximise the effectiveness of existing antibiotics and to prolong their useful lifespan is therefore likely to improve the cost of delivering healthcare (to say nothing of the economic benefit of preventing long hospital admissions due to multi-resistant organisms)

Arguments from defensive practice:

  • Targeted and careful antibiotic therapy using narrow spectrum drugs in short courses is the strategy which minimises both the development of resistance and the risk of harm to the patient from the side-effects of the drug
  • The careful targeting of antibiotic therapy requires a specialist approach, i.e. somebody confident in infectious disease medicine.
  • Statistically, most antibiotics are prescribed by practitioners who do not have extensive infectious disease medicine training, and their prescribing patterns can be described as "defensive", i.e. choosing a broad spectrum agent and using it for longer.
  • It is therefore protective to have guidelines and pathways which support the use of narrow spectrum agents, and restict the use of broad spectrum agents, which the non-specialist prescriber can fall back on.

Elements of inpatient antibiotic stewardship

The CDC document on the core elements of hospital antibiotic stewardship programs  (2019) is probably the best source for this. They list seven points which are easy to remember:

  • Leadership support (adequate resources for the program)
  • Accountability (a physician leader taking responsibility for the program)
  • Pharmacy expertise (multidisciplinary involvement of pharmacy)
  • Action (implementation of auditable interventions)
  • Tracking (monitoring prescribing and resistance patterns)
  • Reporting (feedback to prescribers)
  • Education (of clinicians, about appropriate antibiotic use)

It would be easy to regurgitate this for a paragraph-length answer, such as the first part of Question 2 from the first paper of 2022.  In contrast, the Australian national standard is somewhat less punchy, and revolves around several "quality statements" which basically act as pragmatic recommendations for frontline staff, which are not especially controversial (for example, "take cultures", "document adverse reactions"  and "give antibiotics early for life-threatening sepsis"). ASID, the Australian anagram of the IDSA, do not seem to have specific recommendations for how to pragmatically implement an antimicrobial stewardship program. The best single summary of this is again in Hand (2013), whose Table 3 (Examples of hospital antibiotic stewardship interventions) is so good that it is reproduced below with minimal modification:

Hospital Antibiotic Stewardship Interventions

  • Governance structures:
    • Organisational strategy for antibiotic stewardship
    • Antibiotic stewardship committee (including medical microbiologist or infectious diseases physician, specialist pharmacist and information analyst)
    • Antibiotic prescribing policy (statement of principles of responsible prescribing and expected quality standards), which may include:
      • 48-hour review
      • intravenous-to-oral switch
      • automatic stop orders (termination of prescriptions after a specified interval unless authorisation to continue obtained)
      • compulsory order forms (prescribers required to complete a form with clinical details to justify use of restricted antibiotics)
      • expert approval (prescriptions for restricted antibiotics authorised by infection specialist or head of department)
      • dedicated antibiotic prescription chart
      • removal by restriction (restrictive policy imposed in target wards, units or operating theatres – eg by removing restricted antibiotics from drug cupboards)
      • therapeutic substitution (pharmacists authorised to substitute alternative antibiotics)
      • antibiotic cycling and rotation policy
      • mixing, diversity and heterogeneous prescribing policy
  • Operational delivery
    • Antibiotic formulary
      • may incorporate limited list of antibiotics subject to prescribing restrictions such as requirement for preauthorisation
    • Guidelines for initial treatment of common infections (evidence based, peer reviewed and informed by local resistance data where possible)
    • Guidelines for perioperative prophylaxis for common surgical procedures
    • Reminder systems (eg preprinted adhesive labels for medical case notes)
    • Computerised physician order entry (electronic prescribing) – may incorporate computerised decision-support systems
    • Mobile device software applications for point-of-care information and guidance
  • Risk management
    • Guidelines for management of infection in patients with allergy to antibiotics
    • Information on safe administration of intravenous antibiotics
    • Guidelines for dosing and monitoring of serum levels of toxic antibiotics
  • Microbiology, ID and lab support:
    • Validation and interpretation of microscopy, culture and susceptibility results for laboratory reporting
    • Surveillance and reporting of trends in antibiotic resistance
    • Telephone consultation for advice on infection management
    • Bacteraemia follow-up service
    • Antibiotic stewardship ward rounds
    • Point-of-care rapid tests for bacterial infection
    • Advanced sepsis biomarkers (eg procalcitonin)
  • Controls and quality assurance:
    • Surveillance of antibiotic prescribing trends
    • Public reporting and benchmarking of antibiotic consumption data (eg World Health Organisation-defined daily doses)
    • Audit and feedback of adherence to prescribing policy
    • Audit and feedback of adherence to guidelines
  • Education and training:
    • Induction training on antibiotic stewardship
    • Distribution of printed educational materials (eg pocket guidelines and patient information leaflet)
    • Educational meetings
    • Electronic learning
    • Antibiotic prescribing competency assessment
    • Academic detailing or educational outreach (one-on-one educational intervention)
    • Nominated clinical champions for antibiotic stewardship
    • Provision of patient information and counselling

Now, Question 2 from the first paper of 2022 also asked the candidates to list advantages and disadvantages of this program. This excellent pro/con debate between Phillip George and Andrew Morris (2010) was the main source for what follows, and the figures quoted below originate from their article.

Advantages of antibiotic stewardship in the ICU

  • Antibiotic use and infectious disease is prevalent in the ICU, and critically ill patients are the most vulnerable to adverse drug effects and resistant organisms, so the benefits of antibiotic stewardship would have the greatest impact in that setting.
  • Antibiotics often have serious toxicity, which can be minimised with responsible use
  • Restrictions on the use of some antibiotics may reduce the healthcare costs (the newest most broad-spectrum agents are often the most expensive)- the cost saving is said to be in the realm of US $800,000 per facility per year
  • Multiresistant organisms can colonise ICU patients and increase their length of ICU stay by producing lingering slow-to-resolve illness which increases the cost of healthcare and ICU bed pressure
  • Involvement of infectious diseases physicians in a supervising role may have the inadvertent benefit of improved diagnosis of infectious illnesses (as, in effect, they would be consulted for each one)
  • Involvement of specialist pharmacists could improve the identification of adverse reactions and possible drug interactions, preventing morbidity
  • Use of computer-assisted decision support for antibiotic prescribing could simplify prescribing
  • Standardised order sets could improve empirical coverage, reducing missed infectious agents

Disadvantages of antibiotic stewardship in the ICU

  • There is the possibility that antibiotic stewardship, with its emphasis on stopping or not starting antibiotics, could lead to patient harm from missed sepsis
  • Delay imposed by the need to seek approval could lead to worse patient outcomes (as the timing of early antibiotics is thought to be important for reducing mortality)
  • Infectious disease specialist workload increases dramatically as the result of the increased number of referrals, which has implications for healthcare costs (i.e. the cost savings made through rational antibiotic prescribing are offset by the expense of having to hire a bunch of associate professors for the microbiology lab).
  • The involvement of a supervising infectious diseases specialist is in itself not a guarantee that antibiotic stewardship principles will be followed
  • The implementation of a thorough program requires the sort of resources that would be out of reach for a small regional or rural hospital
  • Antibiotics will still be used, i.e. only the class will change, which means resistance will still emerge - only the pattern will be different.
  • Antibiotic resistance will still develop in the community because of non-medical (eg. agricultural and veterinarian) antibiotic use
  • Antibiotic stewardship may be unacceptable culturally in the ICU, where the intensivist rules as a sort of unquestioned despot, i.e. "you can't tell me what to do, the patient needs horrendomycin because such is my clinical judgment"
  • Looking at the published data, antibiotic stewardship may not improve the clinical outcomes it is intended to improve - few studies report on mortality outcomes, and there does not seem to be any strong evidence that the incidence of adverse drug reactions decreases


Hand, Kieran. "Antibiotic stewardship." Clinical Medicine 13.5 (2013): 499.

George, Philip, and Andrew M. Morris. "Pro/con debate: Should antimicrobial stewardship programs be adopted universally in the intensive care unit?." Critical Care 14.1 (2010): 1-6.