Medication safety - antimicrobial stewardship
What AMS is
- Coordinated system of interventions to optimise which antimicrobial, what dose/route, and for how long - to improve outcomes, reduce toxicity and slow emergence of resistance
- Mandatory in Australian hospitals - a criterion of the NSQHS Preventing and Controlling Infections Standard, plus the AMS Clinical Care Standard
- Antimicrobials are the only drug class where use in one patient harms the next - the ecological argument
The core question set - ask at every antimicrobial decision
- Does this patient actually have a bacterial infection?
- Have cultures been taken before the first dose?
- Is this the narrowest agent that covers the likely pathogen?
- Oral or IV? Is there any reason it cannot be oral?
- What is the stop date? Every antimicrobial order needs one at the time of prescribing
- Has the allergy label been checked and is it real?
Australian context
- Australia has historically had among the highest per-capita community antibiotic use in the OECD
- AURA (Antimicrobial Use and Resistance in Australia) is the national surveillance system; the Sixth AURA report (Australian CDC, Feb 2026) covers 2022-2024
- Reports of critical antimicrobial resistance rose ~25% in 2024
- <50% of surgical prophylaxis is appropriate - the single largest appropriateness gap
- NAPS (National Antimicrobial Prescribing Survey): ~20-25% of hospital prescriptions are inappropriate; the commonest problems are excessive duration and unnecessary broad spectrum
- Aged care: high prevalence of prolonged and prophylactic antimicrobial use, much of it for asymptomatic bacteriuria
- Up to 1 in 5 hospital inpatients is on an antimicrobial at any time
How resistance emerges
- Selection pressure - antimicrobial use eradicates susceptible flora and selects pre-existing resistant organisms
- Resistance is selected for, not induced by, most exposures
- Horizontal gene transfer - plasmids, transposons, integrons; carries multi-drug resistance between species
- Chromosomal mutation - efflux pumps, porin loss, target modification
- Inducible/derepressed AmpC (Enterobacter, Serratia, Citrobacter, Providencia, Morganella) - emerges on treatment with a third-generation cephalosporin
- Collateral damage - disruption of the gut microbiome -> C. difficile, VRE colonisation
Mechanisms by class
| Mechanism | Example |
|---|---|
| Enzymatic inactivation | Beta-lactamases; ESBL, AmpC, carbapenemases (KPC, NDM, OXA-48, IMP) |
| Target alteration | MRSA (mecA -> PBP2a), VRE (vanA/vanB), macrolide erm methylase |
| Reduced uptake / efflux | Pseudomonas porin loss (OprD), tetracycline efflux |
| Target bypass | Trimethoprim/sulfonamide dihydrofolate reductase |
Priority resistant organisms in Australia
- MRSA - including community-onset non-multiresistant strains, high burden in remote Aboriginal and Torres Strait Islander communities
- VRE - Australia has among the highest rates internationally
- ESBL-producing E. coli and Klebsiella - rising in community-onset UTI
- Carbapenemase-producing Enterobacterales (CPE) - IMP-4 endemic in parts of Australia; notifiable
- Multidrug-resistant Pseudomonas, Acinetobacter, azole-resistant Aspergillus, drug-resistant Neisseria gonorrhoeae and Mycoplasma genitalium
Diagnostic stewardship - upstream of prescribing
What AMS measures and audits.
- Take cultures before the first dose wherever it does not delay treatment in sepsis
- *Do not culture what you will not treat*
- No urine culture in asymptomatic patients; asymptomatic bacteriuria is treated only in pregnancy and before urological procedures breaching the mucosa
- No wound swabs from colonised chronic ulcers without clinical infection
- No repeat C. difficile testing as a test of cure
- Use the microbiology properly: Gram stain, susceptibilities, MIC, source control
- Biomarkers: procalcitonin can shorten duration in respiratory infection and sepsis; CRP alone does not distinguish bacterial from viral
Antimicrobial allergy delabelling - a core AMS activity
- ~10% of people carry a penicillin allergy label; >90% are not truly allergic
- Consequence: more vancomycin, more C. difficile, more MRSA/VRE, worse surgical outcomes, higher mortality and cost
- Take a structured history; low-risk labels can be direct oral challenged
Monitoring and audit
- NAPS - hospital, surgical, aged care and quality improvement modules
- Defined daily doses / 1000 occupied bed days; IV-to-oral switch rates; restricted agent approvals
- C. difficile rates, MRSA/VRE bacteraemia rates as outcome measures
- Therapeutic drug monitoring: vancomycin (AUC/MIC-guided dosing preferred over trough), aminoglycosides
1. Prescribing-level - the highest yield
Interventions with evidence, by leverage.
- Empirical therapy from Therapeutic Guidelines: Antibiotic, adjusted to local antibiograms
- De-escalate at 48-72 h once cultures return - the single most examinable AMS action
- Shortest effective duration - most infections are shorter than tradition suggests:
- Uncomplicated cystitis 3 days; CAP 5 days (if afebrile and stable at 48-72 h); cellulitis 5-6 days; uncomplicated Gram-negative bacteraemia 7 days; VAP 7 days
- IV to oral switch as soon as afebrile 24-48 h, clinically improving, absorbing, and an oral agent with adequate bioavailability exists
- High-bioavailability orals: fluoroquinolones, linezolid, metronidazole, fluconazole, doxycycline, trimethoprim
- Surgical prophylaxis: single dose within 60 min of incision; cease within 24 h (ideally at wound closure); redose for long procedures or major blood loss
- The most commonly and most easily corrected inappropriate use in Australian hospitals
- Allergy delabelling
2. System-level
- Formulary restriction and prior approval for broad-spectrum agents (carbapenems, glycopeptides, antipseudomonals, antifungals)
- Prospective audit with feedback - more effective and better accepted than restriction alone
- Automatic stop orders and mandatory indication/duration fields in electronic prescribing
- Clinical decision support, order sets, care bundles
- A multidisciplinary AMS team - ID physician + AMS pharmacist + microbiologist, with executive support and dedicated funding
3. Infection prevention
3. Infection prevention - resistance control is not only prescribing
- Hand hygiene (the single most effective measure against transmission of resistant organisms)
- Contact precautions and screening for MRSA/VRE/CPE; CPE is notifiable in Australia
- Environmental cleaning, device bundles (CLABSI, CAUTI), antimicrobial-free device care
- Vaccination - influenza, pneumococcal, COVID-19 (reduces both infection and antibiotic demand)
4. Community and aged care
- Delayed/back-pocket prescribing for respiratory infection
- Patient education - "antibiotics do not help viral illness and cause harm"
- Aged care: stop long-term prophylaxis, stop treating asymptomatic bacteriuria, review at 48 h
- Real-time feedback of prescribing data to GPs (NPS MedicineWise-style audit and feedback)
Consequences of poor stewardship
- Clostridioides difficile infection - especially with cephalosporins, fluoroquinolones, clindamycin, broad-spectrum penicillins
- MRSA, VRE, ESBL, AmpC, carbapenemase-producing Enterobacterales (IMP-4 in Australia)
- Penicillin allergy labels -> vancomycin use -> VRE and nephrotoxicity
- Antimicrobial toxicity - AKI (vancomycin, aminoglycosides), ototoxicity, QT prolongation (macrolides, fluoroquinolones), tendinopathy and aortic aneurysm (fluoroquinolones), hepatotoxicity
- Intensive care, haematology/oncology, transplantation - highest use and highest resistance
- Aged care facilities, remote Aboriginal and Torres Strait Islander communities
- Agricultural and veterinary antimicrobial use - One Health
- Diagnostic uncertainty and sepsis pathways (time-to-antibiotic targets can drive over-prescribing)
Outlook
- Resistance is not reversible at the population level in the short term - reducing use slows, but does not quickly undo, established resistance
- Where AMS is embedded, Australian hospital data show sustained improvement in appropriateness with continued monitoring - the gains require the ongoing programme, not a one-off intervention
- Global projection: ~10 million deaths/yr attributable to AMR by 2050 on current trajectories
- The development pipeline for new antibacterials is thin; new agents (ceftazidime-avibactam, ceftolozane-tazobactam, meropenem-vaborbactam, cefiderocol) are last-line and are restricted
What to monitor
- Appropriateness (NAPS), total and broad-spectrum usage (DDD/1000 OBD), IV-to-oral switch rate
- C. difficile and resistant-organism bacteraemia rates
- Surgical prophylaxis duration compliance
- Balancing measures - mortality, readmission, length of stay - to show stewardship is not causing harm
🔒
15 more sections, plus exam facts
Premium unlocks every note across every specialty, and the full exam fact library behind it.
Get premium access