Solving For Overprescribed Antibiotics Using BE

There is a general tendency for general practitioners (GPs) to overprescribe antibiotics which can be due to a variety of reasons. GPs may overprescribe antibiotics as they are unsure whether a mild infection may turn more serious. Some can also mistake a viral infection for a bacterial one and prescribe antibiotics, which will not work against viral infections. This has been well documented with respect to conjunctivitis. GPs are often under time pressure, having to see a large number of patients and providing an antibiotic script may be a faster way to complete a consultation rather than explaining why an antibiotic is unnecessary. This also relates to the doctor/patient relationship, where a GP may perceive refusing to prescribe antibiotics as threatening that relationship (“my doctor didn’t do anything for me!”).

These all common, and to some extent, valid reasons for the over prescription of antibiotics. However, such over prescription does come at a cost and this cost is known as antimicrobial resistance (AMR). AMR has been identified as one of the biggest threats to public health today.   

True to its name, AMR occurs when microorganisms, such as bacteria, become resistant to an antimicrobial medicines. Resistant infections are more difficult to treat and, in some cases, untreatable. It can affect anyone, of any age, and in any country.   

The Australian Government found that a lot of Australians were being overprescribed antibiotics during cold and flu season, unnecessarily increasing the risk for AMR. GPs there currently prescribe the greatest portion of antibiotics, out of all health professionals, due to the types of illnesses seen and the large volumes of patients. This makes GPs important partners in the effort to minimise AMR by helping limit community prescribing to only those situations where clinical evidence strongly favours antibiotic prescription.  

What can be done to change prescribing behaviors? Traditional campaigns in healthcare often only rely on education to challenge health related problems. Although well intentioned, these campaigns can be ineffective or even backfire.    
  
So over prescription is a problem. What is the solution?

The Behavioral Economics Team of the Australian Government, also known as BETA, tested the application of behavioral insights to letters sent by the Australian Government’s Chief Medical Officer (CMO). The roll-out of the letters was done through a randomised controlled trial (RCT) involving 6,649 GPs. The inclusion criteria for GPs involved in the study were those whose prescribing rates were in the top 30 percent for their region. Additionally, the letters were sent on 9 June 2017, just before the rapid increase in prescribing which occurs during the cold and flu season.   

The study tested four different letters to GP’s which were designed to cause doctors to reflect on their prescribing behavior in one of four ways - the first three leveraged social proof (the rate at which peers prescribe antibiotics) while including either (1) education about AMR, (2) graphs, or with (3) prescribing materials like respiratory infection action plans. The fourth design was strictly educational regarding AMR.

 The findings were profound – the strictly educational approach, typical of many healthcare campaigns, produced the smallest reduction in prescribing among these practitioners, a 3.2 percent drop.  

Education about AMR alongside peer prescribing data, reduced prescriptions of antibiotics by 9.3 percent.  

Furthermore, peer prescribing data linked with graphical representations produced an even stronger reduction in prescribing, a 12.3 percent drop.  

Thus, letters with behavioral insights or social proof nudges outperformed educational letters by nearly 3-4 times. 
 

The results of this trial demonstrate that a peer prescription behavior letter sent from a respected authority – Australia’s CMO – can have a large impact on reducing antibiotic over prescriptions by GPs.  GPs are busy and need quality data to update their clinical practice while still maintaining their autonomy.  

This study holds promise for the over-prescription of other drugs as well. It highlights the importance of experimentation and testing the efficacy of behaviorally designed messaging relative to simply educational pamphlets as is so commonplace in many sectors beyond health.  Overall, this should inspire governing bodies and policy makers to test similar peer comparison interventions and their impacts over time.  

Read the full study here: Nudge vs Superbugs: A report into a behavioural economics trial to reduce the overprescribing of antibiotics.