This site is intended for health professionals only

Dr Susan Hopkins: ‘We want GPs to educate patients about antibiotics’

The NHS has entered a critical period in the fight against antimicrobial resistance. GPs prescribe 80% of antibiotics in the UK – and have been reducing this activity year-on-year.

But a Government report last year found that despite this success, plus increased public awareness about the dangers of drug resistance, antimicrobial resistance and the incidence of bloodstream infections are on the rise.

It is GPs that seem to be under the greatest scrutiny in their efforts to curb resistance. They have been told that ‘at least’ 20% of all antibiotic prescriptions written in primary care in England are inappropriate – and the Government wants this halved.

Earlier this year it said it wants all prescribing cut by a further 15%.

Speaking to Pulse, PHE’s antibiotics lead Dr Susan Hopkins said she believes GP can continue to go further.

GPs have been successful in reducing antibiotic prescribing so far – why is there not more of a focus on hospitals?

Eighty per cent of antibiotic prescribing occurs in general practice, so clearly where we can make the greatest wins, therefore, is where we have the greatest amount of prescribing.

The other bit is that obviously the people who are rushed to hospital are very sick. So, we’re not really prescribing antibiotics, especially when we’re waiting for the results.

The target in hospitals is really about getting people to review the antibiotic every single day, and only keep going if it’s necessary.

With GPs, what we’re really trying to get them to do is not prescribe, so if patients are clinically well and stable and they believe they don’t need a prescription, to really use that opportunity to educate the patient, and not a prescription.

The other thing we’re trying to do now is push shorter courses. And so in many situations that’s gone from prescribing seven to ten days of antibiotics, right down to five days, and even in some scenarios down to three days, for lower urinary tract infections – cystitis in women, for example.

We understand it’s difficult for GPs – they can’t review those patients every day – but they can prescribe short durations, and if the patient’s not well after three days, it’s about contacting the GP again, and explaining that if this is the case, it’s unlikely that the antibiotic is going to work.

Why are targets still aimed specifically at GPs?

I think GPs feel the target is on them, but the target is very much on secondary care as well. I think the data that we currently hold on antibiotics, in primary care go down to the general practice level, so obviously that’s very granular, so for the 10,000 practices out there, we know exactly what’s happening. At hospitals, we know what’s happening at the hospital level, but at the moment, the prescriptions at the hospital aren’t electronic, so we only know what’s being given to patients at a group level, at a ward level in a hospital. But we are starting to target hospitals more and more, and hospitals have had quality improvement targets for the last three years as well.

What is PHE’s target for GPs to reduce antibiotic prescribing?

We want GPs to reduce antibiotics prescriptions by a further 15% from now, so that’s 2-3% per year – so we’re not asking them to do this in a massive push.

We’ll be measuring the safety of those reductions – ensuring we aren’t seeing lots of repeat consultations for infections at GP practices.

We’ll be making sure that we’re not seeing more admissions to hospital than expected and that we’re not seeing any patients dying in excess of what we already see and expect. That’s really important for us to see and do.

If GPs keep doing what they have been for the last five years, we’ll get there. We’ve already made that change into reducing rather than increasing, and before 2014, which was our highest year for prescribing, antibiotic use in the UK increased by 3-4% annually, and now we’ve reversed that. We’re lower antibiotics users than we were almost a decade ago.

How will the results of these targets be measured?

The key areas we’ve measured in granular detail are from two national GP datasets, the Clinical Practice Research Dataset (CPRD) and The Health Improvement Network (THIN), which code the records of five million patients who consented to anonymous holding of their data.

They come from a variety of general practices scattered all over the country, and with that we can look at every consultation that these patients have, why they’re consulting, why they’ve been using antibiotics.

I think GPs feel the target is on them, but the target is very much on secondary care as well

PHE and Imperial College London looked at the adverse consequences, or any potential adverse consequences, and used those databases with secondary care admissions to see if people are re-consulting.

We can see the number of and the rate of infections and see whether they’re changing; and then we can see whether people who’ve had infections are admitted to hospital or not

How do you know that national campaign messages are reaching patients?

With the Keep Antibiotics Working campaign, which is public-facing and has run for the last two years nationally, we do pre and post tracking with members of the public. We look at what people will think and remember and see from the campaign, and whether it’s changing – and patients report to us that it is changing their behaviour.

GPs also say that it’s supporting them make those difficult decisions.

It’s very difficult when you’re putting out something as a mass media thing, for people to feel it concerns them, until they have that infection.

But when the GP mentions Keep Antibiotics Working, and use the campaign materials that we’ve shared with every GP practice in the country – the non-prescription pads, which allows people to understand the duration of their illnesses without antibiotics and with antibiotics – that gives a GP the opportunity to do that one-to-one education, which is extremely important.

What’s the latest on using point-of-care testing within general practice?

We think that there is evidence for point-of-care tests for C-Reactive Protein, for people who have lower respiratory tract infections and suspected pneumonia, and that clearly has advantages in reducing the need to prescribe.

In respect to other point-of-care tests, there isn’t really a big point-of-care test that we think has made a big difference

Spotting sepsis is going to be a challenge when we’re using more digital technology to do consultations

There has been one looked at for Group-A Strep, but the evidence says that decision-support tools, where you look at how the patient is and the clinical findings on the day of the visit, are just as effective, if not more effective, than point-of-care tests.

We believe we should be using decision-support tools to help GPs, in a risk stratification way, rather than just using a piece of technology that then doesn’t change the prescribing that occurs.

How will you support GP practices to improve their antibiotic prescribing rates?

One of the next steps that we are going to work on with NICE is how we can work with the suppliers for the computers and technology in GP practices – so we can ensure there are toolkits and audits available for them to do within their practice software.

The decision support algorithms that NICE are developing for common infections – those will be available on the computer systems as well. They’re really important next steps, so it makes it easy for GPs to access information at the right time, in the middle of the consultation.

On one hand, GPs are being told to cut their antibiotic prescribing, but on the other there’s a sepsis awareness campaign encouraging the use of antibiotics for ambiguous symptoms. How should GPs navigate this?

The RCGP has developed a set of tools for sepsis that are really good, and clearly it involves seeing the patient. So I think that’s going to be a challenge when we’re using more digital technology to do consultations.

It also involves doing a set of observations on every patient – knowing what their respiratory rate, temperature, heart rate, blood pressure, is. Because they’re the key variables that predict what’s going to happen to the patient in the next 12-24 hours.

If they’re all completely normal and the patient in front of them looks well, with the GP’s clinical experience they can bring those things together, then that’s very reassuring. As long as they document that, then that will help them in any investigations that might subsequently ensue.

Clearly alongside that, there is the safety netting approach, such that when you’re not prescribing, you’re clearly explaining to the patient, or the parents of a child, what the signs are for them to look out for, and where they can get urgent advice 24 hours a day, if anything changes, either through NHS 111, out-of-hours centres, or coming back and calling them again.

So I think it’s really important that we have the dual approach of examining patients, making sure that our set of observations are taken and recorded on the one side, and then on the other side ensuring that people know what to look out for, and I think that’s the standard approach that we should be taking.



  • Trinity College Dublin and London School Hygiene and Tropical Medicine
  • Clinical training in Dublin, Seattle, Paris and London


  • April 2018-Present

Deputy Director National Infection Service, Public Health England

  • January 2007-Present

Consultant in Infectious Diseases and Microbiology, Royal Free London NHS Foundation Trust

Other interests

Taking her son to football matches, walking the dog, yoga and going to music gigs and theatre