How can clinical decision support improve antimicrobial prescribing?

There is still a lot that can be done to improve and optimize the use of currently available antimicrobials. At this year’s Meds Club annual event, Philip Scott, Reader in Health Informatics, University of Portsmouth, will give a talk entitled How can computable knowledge libraries improve clinical decision-support? The case of antimicrobial prescribing. Below is a short discussion with Philip where he explains the challenge of antimicrobial resistance and the current NHS randomised controlled trial which he is helping to conduct in order to measure the impact of a decision-support application for prescribers.

Antimicrobial resistance (AMR) – the resistance of bacteria, parasites, viruses and fungi to medicines – is a global problem. In 2019, a UN Ad hoc Interagency Coordinating Group on Antimicrobial Resistance released a report stating that drug-resistant diseases could cause 10 million deaths each year by 2050, and damage to the economy as catastrophic as the 2008-2009 global financial crisis.

 Antimicrobial resistance is increasing due to the suboptimal use of antimicrobials, and because there is a lack of development for new antibiotics. There is no financial incentive for the Pharma industry to develop new antibiotics, since, if they were invented, and might save lives, they would still be used sparingly, as a last resort in treatment options. In 2019 the WHO identified 32 antibiotics in clinical development that address the WHO’s list of priority pathogens, of which only six were classified as innovative. 

 What’s the challenge with antimicrobial resistance, and how can clinical decision support help? Why is it not helping already?

Antimicrobial resistance is one of the World Health Organization’s priority challenges. There’s a lot of work being done to curb the problem, certainly within the NHS and across the UK. In my presentation, I’ll be reviewing what the current evidence shows about the effectiveness of clinical decision support. And, I’ll be describing a trial that we’re just starting now in which we are actively looking for NHS trusts to participate. Our handheld app aims to provide decision-support modules that will guide prescribers, and we hope to have an impact on antimicrobial prescribing. This is a proper NHS randomised controlled trial that we hope will produce real evidence and will help clinicians see the value of clinical decision support.

Why is antimicrobial resistance still so hard to handle?

There are various reasons why antimicrobial stewardship continues to be a challenge. Sometimes it’s the sheer pace of prescribing. Junior doctors, who are doing the bulk of this work, are faced with the need to move very rapidly from one patient to the next, and perhaps don’t have the Trust’s guidelines or the NICE guidelines at hand, and perhaps they are also facing situations they’re unfamiliar with. To try and get around that, what we’re looking at with clinical decision support is to provide something that’s quick and easy; that doesn’t require a huge amount of time to answer a great deal of questions. It asks a number of simple questions based on the NICE guidelines and local policies so that a very quick recommendation can be given. Or, if it is a complex case, there can be a very clear recommendation suggesting that the junior doctor speak to a microbiologist.

The recommendation system is an app?

In our project, we’re looking to use a mobile app called MicroGuide. It’s currently being used for making simplified versions of trust-prescribing guidance, but in the version that we’re trialling it’s a structured decision-support module. It asks very specific questions about infection, severity, sex, pregnancy, and about infection risks for C. difficile or MRSA, depending on what particular infection is being considered. Its aim is to very rapidly condense the relevant patient parameters so that it can give sensible guidance to the prescribing clinician.

How many clinicians are you planning to involve in the study?

For our trial, we’re not basing it on the number of clinicians, but on the number of hospitals, because the outcome variable that we’re measuring is the volume of carbapenem* prescribing. That’s one of the key indicators which is monitored by what used to be Public Health England. We’re using that as a recognised marker for antimicrobial stewardship, and it is only measured at hospital level. We’re not analysing the behaviour of individual clinicians for the primary outcome, and we’ll be doing qualitative work to understand clinicians’ feelings about using the app, and what’s good and bad about it.

* Carbapenems are a class of highly effective antibiotic agents commonly used for the treatment of severe or high-risk bacterial infections. This class of antibiotics is usually reserved for known or suspected multidrug-resistant (MDR) bacterial infections.

What encouraged you to do this research? What problems, are you aiming at decreasing?

If you look across the whole of the NHS, as is reported on the Public Health England fingertips database, there’s a huge variation in carbapenem prescribing. Now, some of that is legitimate because of the case mix at particular hospitals. But, not all of it can be, so there is a large amount of unexplained variance in the usage of broad-spectrum antibiotics, and that is essentially the problem of antimicrobial stewardship. What we’re saying is, we think that by making it easier for prescribers to understand where a narrow antibiotic, rather than a broad spectrum one, can be used just as well to target a particular infection is a very positive thing, and we hope this study will demonstrate that.

Join us at the Meds Club annual event to hear more.

Learn more about the study by writing to Microguide2.Research@porthosp.nhs.uk.