Antibiotic resistance: what can we do?

When we think of antibiotic resistance, MRSA springs to mind – but the problem is far more widespread, with high levels of resistance in bacteria causing common infections such as urinary tract, respiratory and sexually transmitted infections: but is there anything that we in primary care can do to prevent the arrival of a post-antibiotic era?

As we lurch along in the 21st Century, when the biggest UK killers are cancer and ischaemic heart disease, it is hard to appreciate the role that infectious diseases played before the middle of the 20th Century. Certainly cancer and heart disease existed, but it was infections that frightened people. Industrialisation brought outbreaks of cholera and typhoid to the squalid urban poor. Before that, waves of plague swept the country, and indeed Europe, at regular intervals, with mortality rates that almost defy belief. But perhaps the most telling example of historical killer infections was tuberculosis (TB).

Though accurate records are difficult to acquire, it is believed that TB killed one-third of all those who died in Britain between 1800 and 1850.¹ Furthermore it targeted young adults, people who would otherwise be economically active and bringing up children. Like many infectious diseases, the outlook for succumbing to TB improved markedly with improved social conditions and hygiene, but another significant contribution was made by the discovery of drugs that would kill the causative mycobacterium. This was a triumph for science and social progress: in 1913 there were 120,000 cases of TB notified in England and Wales and 39,000 deaths (from a rather smaller national population, and at a time when notification was probably less complete), but by 2013 there were only 7,400 cases notified and 280 deaths.² However, the 2013 figure is not zero and one important reason for this is the emergence of strains of the TB bacillus which are resistant to multiple antibiotics. If, or more probably when, other seriously nasty infections follow the same pattern (and there is plenty of evidence that this is already happening) we will again face the awful prospect of dealing with waves of incurable infectious diseases. The stakes are really that high.

THE SCALE OF THE PROBLEM

It has been apparent for some years that resistance to the antibiotics generally used in the treatment of bacteria is becoming more common (it is, of course, the bacteria which acquire resistance, not individual patients). Public Health England estimates that by 2050, antimicrobial resistance (AMR) will cost the global economy $100 trillion. This also translates into 10 million deaths globally.³ AMR is truly an international concern as infectious diseases are these days concentrated in less economically developed countries, countries that also are going to be less able to pay the costs. The World Health Organization⁴ is deeply concerned about AMR, stating: ‘Without urgent, coordinated action, the world is heading towards a post-antibiotic era, in which common infections and minor injuries, which have been treatable for decades, can once again kill.’ (Box 1) (Pedantic note: antimicrobial resistance refers to resistance to antibiotics, but also to anti-fungal and anti-viral drugs. As antibiotics are used much more in primary care than the other two categories, this article focuses on antibiotics).

At present it is believed that the inappropriate use of antibiotics in humans contributes to AMR. Accordingly much of the focus of dealing with AMR is falling on the healthcare workers who prescribe the antibiotics, and the patients who take them. The first World Antibiotic Awareness Week ran from 16 – 22 November 2015, and for several years there has been a European Antibiotic Awareness Day which last year was 18 November. Your practice may have been involved with the campaign. My practice Patient Participation Group is very concerned about AMR and members were buzzing around the waiting room that day handing out leaflets: our patients are also aware of the implications of doing nothing.

WHY IS THERE A PROBLEM?

It would be tempting to blame the pesky bacteria for developing resistance. However they are just following their own Darwinian destiny by trying to survive. There are several other factors to consider, all of which contribute to the problem: addressing just one factor and ignoring the rest is unlikely to bring about the desired result.

1. Over-prescription of antibiotics

There is an almost universal consensus that AMR would be ameliorated if antibiotics were prescribed in a more logical way, a view supported by most major UK medical authorities.⁵ This makes AMR firmly a general practice problem as 80% of all the antibiotics prescribed in the UK are prescribed from primary care.⁶ What this boils down to is: don’t recommend antibiotics if the infection is likely to be viral. The logic is that if there is more antibiotic floating around in the environment, then there is more chance of the drugs coming in contact with a passing bacterium, and thus starting the process that results in AMR.

2. Other antibiotic use

Veterinary practitioners use almost the same range of antibiotics as are used in humans. According to the Soil Association, about 45% of antibiotic use in the UK is for animals⁷ (in the USA it is 72%⁸). Sometimes this is to treat sickness in the animals, and sometimes as prophylaxis to prevent infection. The use of antibiotics in farming is also known to increase weight gain, a practice outlawed in the EU since 2006, but still used in other parts of the world. Use in farming poses a risk to the antibiotic-allergic human who may be unwittingly consuming antibiotics in their burgers. It also adds to the antibiotics in the environment, and so risks increasing AMR. Farmers are also being targeted with advice to reduce antibiotic use.⁸

3. Antibiotic development

There have been no discoveries of new antibiotic groups in the last 30 years.⁷ This is not to say that drug companies are not churning out more products – a record number of new drugs were authorised in 2014, and between 1990 and 2010 the number of available drugs grew by a factor of 2.5. It may be that all possible antibiotics have already been discovered, but this seems unlikely and rather pessimistic. Because of the current patent and marketing regulations, it makes more sense for a company that wishes to maximise profits to produce variants of old drugs rather than innovative new ones which may not acquire a market. Drug companies spend roughly twice as much on marketing as they do on research and development.⁹ Of course, other people are carrying out research into antibiotics – for example charities, universities, and the NHS itself – but the resources employed are dwarfed by the research going on in the pharmaceutical industry.

WHAT IS TO BE DONE?

If general practice uses antibiotics more rationally, will this reduce AMR, or at least delay the progress of AMR? NICE published guidance this year, making various recommendations and offering evidence for them.¹⁰ I have been through the 69 papers identified as supporting evidence (so that you don’t have to) and I regret to report that I could find not a single piece of evidence that answers our central question. Using antibiotics more rationally is certainly a plausible way of cutting the quantity that is prescribed. Rational use can also be supported on the grounds of economy and minimising possible adverse reactions to the drugs. In addition, rational use should shift the emphasis away from prescribing an antibiotic (which won’t work) and towards making suggestions about other supportive treatments that might work. However to my mind the case has not yet been proved that getting GPs to prescribe antibiotics according to the guidelines will reduce AMR.

The guidance

The NICE guidance¹⁰ makes several recommendations for prescribers. Below is a selection for consideration (for the full list, check out the reference). Those of you who are operating triage systems or minor illness clinics that may include prescribing might like to double check to make sure your protocols are consistent with this guidance.

1. ‘When prescribing antimicrobials, prescribers should follow local (where available) or national guidelines on prescribing:

• The shortest effective course
• The most appropriate dose
• The route of administration’

These suggestions are hard to disagree with. Most areas have their own local formulary (often compiled by secondary care), as do some individual practices. In addition, practice computer software can be configured to preferentially select medicines that appear in a formulary, and indeed some software can be quite rude to you if you try and prescribe something different. Failing this, the British National Formulary remains a font of prescribing wisdom and advice.

The chosen dose of antibiotic is something else that a formulary will help with. Special care needs to be taken with children and the elderly (who tend to need lower doses). Penicillin allergy is fairly common, and allergies to different antibiotics may also occur. In addition, some people are intolerant of some antibiotics – for example, experiencing diarrhoea when taking amoxicillin. The choice of antibiotic may also depend on any other medicines being taken: the practice prescribing software will alert you if you try to prescribe something inadvisable.

When an antibiotic is being tested to see if it works, the length of a course is chosen by inspired guesswork, but basically at random. If the trial is completed and the course works, then that becomes the standard recommended course. This is why nearly all antibiotics suggest a 5-day or a 7-day course – these are the courses that have been used in the clinical trials. Shorter courses may well be successful, such as a 3 day course of trimethoprim in uncomplicated lower urinary tract infection. In any event, some patients will only take their antibiotic for as long as they feel unwell and then stop. There are several possible reasons for this behaviour: some forget to take the antibiotic and do not have the symptoms to prompt them; some stop because they think this is the right thing to do; and some stop because they want some antibiotics in stock for future use. This habit of not completing courses of antibiotic could well have an impact on AMR (it certainly seems to have been a factor in the emergence of tuberculosis resistance, where the recommended courses are often several months long).

2. ‘For patients in primary care who have recurrent or persistent infections, consider taking microbiological samples when prescribing an antimicrobial and review the prescription when the results are available. For patients who have non severe infections, consider taking microbiological samples before making a decision about prescribing an antimicrobial, providing it is safe to withhold treatment until the results are available. Consider point of care testing in primary care for patients with suspected lower respiratory tract infections’.

The reference to lower respiratory tract infections is all about getting some blood for a C-reactive protein level (CRP) before prescribing an antibiotic. Where CRP point of care testing is available, it has been shown to reduce unnecessary antibiotic prescribing by accurately identifying patients who do and do not require antibiotics for respiratory symptoms. Where CRP testing is not available at the point of care, the feasibility of testing depends on how quickly the samples can be obtained and the results received. In my locality the local hospital will take bloods the same day and send a result within a few hours. This is however scarcely practicable for a patient with pneumonia who feels dreadful and can’t get to hospital. Taking blood at the surgery takes a lot longer. Culturing microbiological specimens takes at best a few days, during which delay treatment could have started. And then there has to be an additional contact with your patient to relay the result and suggest treatment. So investigation increases workload and may disadvantage patients. GPs have a fine tradition of predicting the nature of infections on the spot and prescribing accordingly. Might it just be sensible to let them get on with doing what they do best?

3. ‘Prescribers should take time to discuss with the patient and/or their family members or carers (as appropriate):

• The likely nature of the condition
• Why prescribing an antimicrobial may not be the best option
• Alternative options to prescribing an antimicrobial
• Their views on antimicrobials, taking into account their priorities or concerns for their current illness and whether they want or expect an antimicrobial
• The benefits and harms of immediate antimicrobial prescribing
• What they should do if their condition deteriorates (safety netting advice) or they have problems as a result of treatment
• Whether they need any written information about their medicines and any possible outcomes’.

Whether an antibiotic is prescribed or not, this is a bare minimum that a consultation for an infection should cover. The use of antibiotics is only one part of the management of infectious disease. Importantly the guidance also covers the possibility of a delayed prescription for antibiotics, to be used only if spontaneous improvement does not occur. Many of our patients are aware of the problem of AMR, so it is not justified to assume that they are all after antibiotics and that every consultation will end up being a battle round this one issue. That said half of patients who see their GP with a respiratory infection expect to be prescribed an antibiotic.¹¹ A quarter of folk in the UK believe that antibiotics cure colds and flu.¹² Patients denied antibiotics by their GP are more likely to be disgruntled.¹³ So it may be that you and your patient will not be in agreement on antibiotic prescribing: discussions are likely to be difficult. Your time is important and there must be a point at which it is legitimate to give up and issue the antibiotics, in the full knowledge that they probably won’t make any difference. Is it worse to issue one unnecessary antibiotic prescription than to break your promises to 15 other patients who have been given appointment times?

Our patients are, in general, rational beings. They will alter their health beliefs if the reasons offered are good enough. The only reason to want antibiotics is because you believe they will make you better. If they are unlikely to do this, then the balance with the disadvantages (possible adverse reactions, prescription charges and – and this is a surprisingly convincing point for many patients – cost to the NHS) shifts strongly towards not using the antibiotics. Some infections are usually bacterial, some are usually viral, but there are a lot that can be either bacterial or viral, and your patient may want to take their chances and use the antibiotics just in case. Reaching a shared understanding of the problem is not just about convincing your patient that you are right, it is about negotiation and the possibility of mutual compromise.

This recommendation also supports a point made earlier: be sure that your patients get all the relevant advice about their infection, to be used whether or not they also get their ‘magic’ pills. Even if they do work, antibiotics often take 48 hours or more to start to take effect. Symptomatic advice – paracetamol, fluids, and rest – are also important and should not be forgotten while you are bickering over a prescription. Also, bacterial infections are often a complication of viral infections so things may go wrong with your management plan. The illness starts viral but becomes bacterial. In situations like this, safety-netting is very important: describe what you expect will happen, and make stipulations about when a further consultation is needed. For example, with a respiratory infection if your patient becomes unaccustomedly breathless, or starts coughing up blood, or becomes more unwell, then this should prompt another discussion with a healthcare professional.

CONCLUSION

AMR is an escalating problem, the impact of which is likely to be felt more in the poorer countries of the world, and people are rightly concerned about it. It is possible that the prescribing habits of general practices in the UK may be contributing to the problem, but this is only one of several important factors which should all be addressed. Evidence that beating up primary care workers for their antibiotic prescribing will solve the problem is lacking.

However. it should be a matter of professional pride that general practice uses antibiotics in a safe and evidence-based way. If the cause of a sore throat is viral,14 then giving antibiotics can only cause harm because it is impossible for it to do any good.

REFERENCES

1. C N Trueman. Diseases in industrial cities in the Industrial Revolution.

The History Learning Site, 31 Mar 2015. http://www.historylearningsite.co.uk/britain-1700-to-1900/industrial-revolution/diseases-in-industrial-cities-in-the-industrial-revolution/

2. Public Health England. Tuberculosis (TB): annual notification and mortality data (1913 onwards). October 2015

3. Public Health England. New report shows stark effect of antibiotic resistance. December 2014

4. World Health Organization. Antimicrobial resistance. Fact sheet 194,

April 2015

5. Anon. Calls to prevent antibiotic misuse. BMA News 5 December 2015 p2.

6. RCGP. TARGET Antibiotic Toolkit. www.rcgp.org.uk › Clinical › Toolkits

7. Soil Association. Antibiotic resistance: a looming crisis. SoilAssociation.org

8. Gallagher J. Farmers urged to cut antibiotic use.BBC News website. www.bbc.co.uk/news/health-35030262

9. Naci H, Carter A & Mossialos E. Why the drug pipeline is not delivering. BMJ 2015;351:h5542

10. NICE. Antibacterial stewardship: systems and processes for effective antimicrobial medicine use. NG15 August 2015 https://www.nice.org.uk/guidance/ng15

11. McNulty CAM et al. Expectations for consultations and antibiotics for respiratory tract infection in primary care. Br J Gen Pract 2013;63:349-350.

12. Anon. Myth that antibiotics cure coughs and colds still rife. BBC News website 18 November 2011. www.bbc.co.uk/news/health-15772727

13. Ashworth M et al. Antibiotic prescribing and patient satisfaction in primary care in England: cross-sectional analysis of national patient survey data and prescribing data. Br J Gen Pract 2016; DOI: 10.3399/bjgp15X688105

14. NICE CKS. Sore throat – acute. cks.nice.org.uk/sore-throat-acute