TRAINEE GPs are just as prone to overprescribing antibiotics as their older peers, according to new Australian research.

The first large-scale study of its kind, published in Family Practice looked at the prescribing habits of nearly 900 GP registrars in relation to cases of acute otitis media and acute sinusitis – two conditions where current guidelines frown on the routine prescribing of antibiotics.

Across 100 000 consultations, identified in a cross-sectional analysis of data from the Registrar Clinical Encounters in Training (ReCEnT) study, an ongoing, multicentre prospective cohort study, one of these conditions was diagnosed in around 3000 cases. Antibiotics – mostly amoxicillin – were prescribed for 79% of cases of otitis media and for 71% of cases of sinusitis.

These prescribing rates are far above the 20% benchmark expected if the participating registrars had been prescribing according to evidence-based guidelines, said the study authors from the University of Queensland and other institutions.

While it’s true that junior doctors were prescribing antibiotics a little less than their older colleagues, the difference was not great, suggesting that current training programs were not doing enough to deal with the problem.

“The findings are not that surprising, since it’s established that GPs prescribe highly,” said lead author Dr Anthea Dallas, a Sydney-based GP and researcher at the University of Notre Dame School of Medicine. “But it’s still disappointing, as you’d hope registrars would prescribe less, and that they’d be educated to do so, but it doesn’t seem to be the case.”

Dr Dallas said the reasons why trainee GPs may struggle to say no to antibiotics were complex and multifactorial.

She said one factor was that junior doctors were transitioning from working in hospitals, where infections tended to be more serious and antibiotics were prescribed more liberally than in the community.

“It could be hard for trainees to adjust their mindset from hospital emergencies to community medicine,” Dr Dallas said.

Another problem was diagnostic uncertainty.

“When you don’t have much experience, you tend to prescribe on the side of caution. You may also be more likely to ask advice from your supervisor, who would have started practising in an era when antibiotics were prescribed more freely. And, in fact, our study found that prescribing antibiotics was associated with seeking advice.”

Communication skills also played a role: doctors needed to know the evidence and clearly explain it, but they should also be able to properly understand patients’ needs.

“It may be that the doctor thinks there’s pressure from the patient to prescribe antibiotics, when in fact there isn’t,” Dr Dallas noted.

One finding from the study was that antibiotic prescribing for otitis media was no greater in younger age groups. It was the one area where you may expect to find higher prescribing, since Australian guidelines recommend antibiotics for infants under 6 months of age and state that they “should be considered” in children aged up to 2 years.

This suggested that trainees may be finding the guidelines too complex and confusing to use on a day-to-day basis.

But the biggest surprise was that the registrars who prescribed the most were those in their later stages of training, at the point where they should have developed more confidence in diagnosing and treating their patients.

This again pointed to insufficient training and guidance in this critical aspect of clinical practice.

Dr Dallas said that the problem of overprescribing antibiotics in primary care was a huge concern in antimicrobial resistance because it was GPs who were responsible for the bulk of the prescribing.

“We need to target junior doctors because they’re the future of prescribing. And it’s not that they’re unaware of the resistance problem, it’s that the daily pressures often override the theory.”

She said that another reason to target trainees was that they were at the stage of developing their habits, but those habits were not yet set.

What was needed was targeted educational interventions to help junior doctors transition from hospital to community practice, the study authors wrote. And those interventions needed to address not just the junior doctors but their supervisors as well.

Training should cover the use of guidelines, communications skills and strategies to deal with diagnostic uncertainty.

The study authors said that they had already piloted some modules for improving prescribing for respiratory tract infections, which could be adapted for other indications.


To find a doctor, or a job, to use GP Desktop and Doctors Health, book and track your CPD, and buy textbooks and guidelines, visit doctorportal.




A public health campaign is needed to educate patients about appropriate antibiotic use
  • Strongly agree (81%, 155 Votes)
  • Agree (12%, 23 Votes)
  • Disagree (4%, 7 Votes)
  • Neutral (2%, 4 Votes)
  • Strongly disagree (2%, 3 Votes)

Total Voters: 192

Loading ... Loading ...

2 thoughts on “Antibiotic overuse: junior doctors are at it too

  1. Ian Hargreaves says:

    Having been sued a couple of years ago regarding a post-operative infection, I was bewildered/surprised/dismayed to find that antibiotic stewardship has no legal status. Talking to both the defence union lawyer and the College of Surgeons representative, the only relevant question in NSW law is that when a complication develops, would a particular treatment have possibly benefitted this particular patient. Population values are not part of the equation – the question is not whether you are generally a safe driver, but whether you ran over this individual.

    Although the textbook chapter on infection in Green’s Operative Hand Surgery (written by David Green himself) stated that antibiotics were not necessary for this procedure, it turns out also that this is inadmissible as evidence, unless one can actually get the international expert to testify in the NSW Court. Neither was the Cochrane Collaboration evidence (in my favour) of any use to my defence team.

    While this case was eventually successfully defended, it is certainly easier and a lot less stressful to prescribe unnecessary antibiotics for thousands of patients, than to get sued once every 20 years. Perhaps the reason that more senior trainees in this study were more likely to give antibiotics, was that they realised the consequences of failing to treat something like a single meningococcal infection in a timely fashion.

    Funnily enough, the experts who write the antibiotic guidelines are remarkably reluctant to present themselves for cross examination in a Court, in relation to an individual case. My lawyer tried unsuccessfully to get local infectious disease experts to write a report or testify.

    If we want to take stewardship seriously, there must be legislation (akin to the Good Samaritan cover for doctors) preventing AHPRA complaints or lawsuits relating to compliance with prescribing guidelines. Or perhaps the authors of this study (or the authors of the guidelines) would agree to indemnify any young doctor who is accused of not prescribing necessary antibiotics for a child who dies of meningitis after a presentation misdiagnosed as otitis media.

    In each individual patient contact, the thoughtful doctor weighs up the potential benefits and risks of antibiotic therapy, compared with the risks of missing a catastrophic infection. It is no surprise that the vast majority err on the side of caution rather than following the advice of Ivory Tower academics. I would hate to be a young doctor seeing 100 sick children, believing that 70% needed antibiotics, but being bullied by my superiors to only prescribe antibiotics to 20 or fewer of my patients, at the risk of my career progression.

  2. Jason Delaney says:

    The ReCEnT study is recording data from clinical encounters (like Beach used to do) – It is not investigating if a specific patient meets accepted criteria for needing antibiotics or not. It does not look at if the registrar sees a well child with an URTI and slight red ear is coding that as an URTI (which what I tend to do theses days) rather than Acute Otitis media and not prescribing to that patient group. It does not record how long the patient has been sick or presence of systemic features which may make antibiotics warranted. It is not clear how you have come up with the 20% benchmark. Do training practices whom take registrars have the same population of patients as general practice in general? I call 20 years ago patients would present with the sore ear that started in the middle of the previous night. These day my patient group tends to present more with sore ear that has been going on for 2 days and not getting better – potentially a true indication for antibiotics.
    ReCEnT collects great useful data but are you starting make interpretation beyond the data you have collected?

Leave a Reply

Your email address will not be published.