Why GPs prescribe too many antibiotics and why it's time to set targets

Published yesterday, our study estimates Australian GPs are prescribing about five million too many scripts for antibiotics a year for run-of-the-mill respiratory infections. But this is not a simple case of “blame the GP”.

What our study does show is many years of educating GPs about appropriate antibiotic prescribing, and the link to antimicrobial resistance, has failed to stem over-prescribing.

So, it’s time to set a national target for antibiotic prescribing in general practice, just like we set targets for carbon dioxide emissions to control the effects of climate change. Local Primary Health Networks could support GPs to meet these targets.

We’d also need to support GPs to easily and cheaply acquire the skills to help them reduce their prescribing safely. There are already moves towards supporting GPs this way. However, we should be prepared for a slow and sustained effort.

If GPs can’t make these changes, they risk more draconian measures being imposed on them by government or bodies like the Australian Commission on Safety and Quality in Health Care. This might include GPs needing to seek an Authority Prescription from the Pharmaceutical Benefits Scheme to prescribe some antibiotics, and punitive measures being imposed on those prescribing beyond some arbitrary limits.

What did we do and what did we find?

We looked at the actual rates of antibiotic prescribing for acute respiratory infections, like sore throats, acute coughs (also called acute bronchitis), and acute middle ear infections. Our data was collected by a survey of about 500 GPs from across Australia, who recorded what they did in every consultation for two weeks.

We then compared that with the rate that would have occurred had every GP stuck rigidly to Therapeutic Guidelines, highly respected national prescribing guidelines many GPs use.

While we had expected about half of actual prescribing to meet the guidelines, we found just 11-23% met them. In national terms, that’s almost six million antibiotic prescriptions a year for these acute respiratory infections, compared with around one million a year had GPs stuck to the guidelines, a difference of roughly five million prescriptions a year.

Why is this important?

Each course of antibiotics contains roughly five grams of antibiotics. So, if GPs had stuck to the guidelines, we could safely reduce antibiotics use by 25 tonnes a year.

This mound of antibiotic represents an aspiration – what we could avoid, with minimal harm to the Australian public, and enormous benefits to reducing the generation of community acquired antibiotic resistance.

Further reading: We know why bacteria become resistant to antibiotics, but how does this actually happen?

In the past we have not really had any target to aim for, but instead wondered if we should aim for the rates achieved by other countries such as the Netherlands (about half of our rates).

Our data show we could take that target much further.

Why do GPs prescribe too many antibiotics?

There are many reasons GPs prescribe too many antibiotics. GPs (and their patients) might want to minimise the risk of their patients being exposed to a dangerous bacterial infection that might have been avoided by prescribing antibiotics early.

Then there’s the diagnostic uncertainty that bedevils this part of primary care. Every apparently trivial cough or cold a GP sees could be the early stages of a dangerously serious infection, like community acquired pneumonia, meningitis, or quinsy (a complication of tonsilitis), and it is often very difficult to be sure in a single visit.

Improving diagnosis might be possible using near-patient testing – a quick test in the surgery, rather than sending off a sample to a laboratory for testing. But these tests are only partly satisfactory because they are not always accurate enough, and they are very expensive, perhaps doubling the cost of the consultation.

Other important factors are:

• pressure from patients for GPs to prescribe antibiotics, either real or supposed by the GP. GPs often say this is a major influence, but other studies say it is often over-estimated by GPs

• an assumption the consultation will be over quicker with a terminating prescription in time-poor general practice

• commercial anxieties (“if I don’t give the patients what they’ve come for, they might go to other GPs more willing”)

• habit (“why change what’s been working just fine 10 or 20 years ago if it isn’t broke?”), remembering that the consequences of antibiotic resistance happen in hospital care, far removed from this patient now

• “failure of the commons”, in which a shared resource (in this case the absence of antibiotic resistance) is threatened by many individual interests (the individual is sick and wants whatever might quickest make them feel well again).

What needs to happen?

It’s easy to jump to the conclusion from our findings that GPs should “stick to guidelines” when it comes to prescribing antibiotics. But that’s unrealistic. Guidelines are no more than their name suggests, simply a guide to how to manage a patient and their illness.

The real world is much more complicated: patients have additional illnesses, and other demands (often social, psychological or even just preference – for example, avoiding the risks of some symptoms even at the expense of some harms) – and the skillful GP needs to balance all this.

Our results, which demonstrate higher than expected rates of excess antibiotics prescribed, means we have a lot of antibiotic savings we could safely make.