Dr James Larcombe describes how an evidence-based antibiotic policy developed by his practice led to a fall in prescribing rates and increased patient satisfaction

Serious problems with resistant bacteria are now a feature of community as well as hospital practice.

An average GP writes 1800 prescriptions for antibiotics a year; 900 of these are for respiratory infections and 300 for urinary infections.1 GPs account for 80% of antibiotic prescriptions by doctors and 40% of all antibiotics prescribed in the UK (50% come from veterinary and agricultural practice).1

Considerable variations in prescribing habits exist,1 but prescribing habits are amenable to change.2 Thus GPs are in the front line of the battle against antibiotic resistance.

As a group, we have had a particular interest in managing infections and making efficient use of antibiotics for many years. We don't really know how this came about: probably a meeting of like minds, rather than a conscious decision to target this area of practice.

Efficient prescribing has been a key issue for us. Our first practice formulary was developed in the early 1990s, building on a number of informal initiatives.

As partners and registrars have joined us, they have taken these interests on board, bringing their own thoughts, but enhancing rather than diminishing our efforts.

The use of antibiotics is the most malleable of all prescribing decisions: a large percentage of prescribing is acute, is particularly subject to doctor3 and patient expectation,1,4 and is generally, though not universally, primary care led.1

The title of our submission was 'An evidence-based, patient-centred, holistic approach to the treatment and management of infections in general practice'. Central to our submission were two major efforts to improve practice:

1. A comprehensive appraisal of the investigation, management and treatment of urinary tract infections (UTIs), performed as a Diploma project.

This incorporated a thorough appraisal of evidence, seeking views on diagnosis, management and best practice, and guideline development by all nurses and doctors following thorough discussion. While this may seem a bit excessive, it has surpassed our expectations in terms of reducing workload, and improving efficiency, team working, and patient care.

2. Discussion of the management of upper respiratory (and related) infections.

One of our partners had undertaken a thorough search of the evidence during MRCGP revision. This was presented to the doctors, discussed, and an informal protocol for the management of these common conditions was agreed.

One of the most important steps was the introduction of delayed prescriptions,5 which allows the reduction in antibiotic prescribing at individual rates without destroying the concept of a cohesive policy, and better meets the expectations of patients.

We felt that optimising antibiotic prescribing could not be seen in isolation and identified six elements to improving our prescribing behaviour:

  • Maximising non-antibiotic treatments
  • Maximising preventive treatments
  • Maximising immunisation
  • Efficient use of investigations
  • Improving teamwork in the management of infections
  • Improved consultation skills (especially by increasing awareness of patients' reasons for attendance).

This approach is multifaceted and complex. Starting from scratch, practices would find it difficult to implement a replica in a short period of time, but most people already practise a number of the elements we encourage, perhaps without realising it.

To help retain the key messages, we produced a template (see Table 1, below) which set these elements (condensed into 'Interventions', 'Non-antibiotic treatments', and 'Antibiotic treatment: type/length) against our eight targeted infections.

Table 1: Beacon practice 99: Summary of interventions
Beacon practice 99: summary of interventions

There are two approaches with proven effectiveness that we haven't pursued to any great extent. One is providing written information to patients;6 the other is computer-aided prescribing.

There are many good examples of patient leaflets, and most are available on the internet. We have concentrated on a personal approach within the consultation, but acknowledge that the use of handouts would complement and enhance our information.

Computer-assisted prescribing has been specifically tested with regard to antibiotic prescribing in hospital practice in the USA.7 A primary care system in the UK – PRODIGY (www.prodigy.nhs.uk) – seems likely to fulfil this role.

Antibiotics need to be given in adequate doses as well as at the right time.1 A blanket, unquestioning approach that treats serious problems the same as simple problems has its own dangers.1 Reflecting this view we encourage effective prescribing:

  • Choose the right antibiotic
  • Choose an appropriate dose
  • Choose an appropriate duration.

We favour bactericidal antibiotics (e.g.ipenicillins) for severe infections, and drugs with superior in-vivo activity (e.g.trimethoprim in UTIs).8 In UTI, it has been shown that apparent (in-vitro) levels of resistance do not accurately predict therapeutic success.9

We tried to be pragmatic and took opportunities where they presented. As much was achieved by the spoken word as by the written word. When anyone showed a spark of interest we encouraged them to formulate ideas and generate supportive evidence. This enhanced debate, and appears to have led to genuine change.

The levels of robustness of the evidence collected does vary (see Table 2,below). Six topics received extensive scrutiny from high quality external reviews. Our sources were: Clinical Evidence, The Cochrane Library and Best Evidence (both available on CD), and MEDLINE searches, as well as local and regional evidence-based guidelines.

Table 2: Robustness of the evidence

Infectious conditions receiving a comprehensive evidence-based approach Infectious conditions receiving an intermediate evidence-based approach

Sore throat
Earache (otitis media)
Urinary tract infection
Helicobacter pylori

Earache (otitis externa)
Head lice
Vaginal infections
Immunisation policy

Evidence relating to another eight topics was gathered from multiple but less rigorous sources, e.g. Drug and Therapeutics Bulletin (DTB) and MeReC issues, recent peer-reviewed papers, as well as high quality guidelines.

We also considered a few other conditions where we relied on single sources of information only, e.g. the drug management of herpes zoster (based on a DTB article), or methicillin-resistant Staphylococcus aureus (MRSA) (based on a health authority guideline).

Are patients any better for the guidelines? Measuring how our patients have fared individually is difficult. Since we are encouraging the withholding of (unnecessary) treatment, a positive outcome will often be a lack of apparent harm rather than an obvious improvement in health.

Patients are likely to suffer less iatrogenic disease as a result of lower prescribing.10 Referrals for UTI in children have increased from 19 in the 3 years before our guideline setting exercise, to 32 in the three subsequent years. It is most likely that this represents an improvement in diagnosis rather than an improvement in management and referral behaviour.

Influenza vaccination in autumn 2000 and a pneumococcal campaign in 1999 achieved rates of cover of high-risk patients in excess of 70%.

A recent audit identified no deterioration in the rates of admissions for pneumonia and UTIs/pyelonephritis before and after the changes encouraged by our evidence-based reviews.

Review of our admissions book showed that over the past 4 years benzylpenicillin had been administered before hospital admission in all patients where meningitis or meningococcal septicaemia was considered a possibility.

Our pattern of antibiotic sensitivity remains unchanged over the past 5 years according to local laboratory results. This is not as encouraging as one published study,11 but our prescribing has been subjected to fine-tuning rather than a major shift in practice.

Our approach has been labelled 'patient-friendly'. It encourages partners to talk about the reasons for prescribing decisions and involves patients in the decision to treat. We have had no complaints about non-prescription of antibiotics.

Partners are much happier about saying 'No' to patients who request antibiotics. This seems to be due to three factors:

  • A greater appreciation of the problems as well as the benefits of antibiotics by patients
  • The flexibility engendered by listening to patients and choosing the most appropriate treatment for them
  • The reassurance from good quality evidence and its widespread endorsement.

Within our practice

Review of our performance 3 years after implementation of the UTI guideline revealed continuing satisfaction with the messages it conveyed to the doctors and nurses in the practice.

The review also showed continuing success in implementation of a low investigation policy (see Table 3,below), and an improved rate of 3-day trimethoprim courses (60% of all trimethoprim is currently prescribed as a 3-day course, compared with only 5% before guideline implementation).

Table 3: Average number of mid-stream urine specimens requested per month, 1997 to 1999

  Pre-project 1997 Post-project 1997 1999
Children 14 13 8
Adult males 11 5 5
Adult females 48 21 25
Elderly 36 14 8
Totals 109 53 46

Interesting results were achieved by different methods of implementing prescribing change (Figure 1, below). Between 1995 and 1997 we actively promoted the use of formulary antibiotics (see Figure 2, below). Between 1997 and 1999 we actively implemented evidence-based prescribing.

Figure 1: Changes in antibiotic prescribing following the introduction of a formulary and evidence-based practice
table and graph


Figure 2: Practice policy on antibiotic use
practice policy

The former resulted in a change in the type of antibiotic prescribed, but not the amounts; the latter showed a dramatic fall in the overall prescribing rate and a drop in the prescription of formulary drugs used in respiratory infections.

Our overall prescribing of antibiotics continues to fall and remains the lowest in our PCG. Quality markers at PCG level are all met. Our most recent quarterly PACT report, which assessed antibiotic prescribing, made satisfactory reading.

We are about to embark on a sophisticated audit of the effectiveness of our use of delayed prescriptions. We will look at how often they are used, and whether the numbers of prescriptions actually cashed is similar to the figures suggested by Little et al.5

The whole purpose of the Beacon Practice award is to 'spread the word'. As a pilot Beacon practice, we began our 'mission' in September 1999.

We held seminars on our surgery premises on two afternoons. Delegates from general practice, PCGs, and health authorities, as well as microbiologists from the Public Health Laboratory Service, attended.

We have given hour-long presentations at Beacon and 'Prescribing' Fairs in Harrogate, North Yorks, and have attended lunch-time lectures, and a Conference on Quality in General Practice.

We have spoken to influential committees, including the Standing Medical Advisory Committee (SMAC), and have shared thoughts with the National Prescribing Centre, who have produced a guide to implementing prescribing change within PCGs. In addition, we have received numerous requests for further information.

This article cannot convey the totality of our work, and anyone interested in further infomation may wish to look on our website www.doctorssedgefield.co.uk (currently under construction) or contact us directly.

  1. Standing Medical Advisory Committee Sub-group on Antimicrobial Resistance. The Path of Least Resistance. London: DoH, 1998.
  2. Thomson MA, O'Brien, Oxman AD et al. Educational outreach visits: effects on professional practice and health care outcomes (Cochrane Review). In: The Cochrane Library 2000, Issue 3. Oxford: Update Software.
  3. Britten N, Ukoumunne O. The influence of patients' hopes of receiving a prescription on doctors' perceptions and the decision to prescribe: a questionnaire survey. Br Med J 1997; 315: 1506-10.
  4. MacFarlane J, Holmes W, MacFarlane R, Britten N. Influence of patients' expectations on antibiotic management of acute lower respiratory tract illness in general practice: questionnaire study. Br Med J 1997; 315: 1211-14.
  5. Little P, Williamson I, Warner G, Gould C, Gantley M, Kinmonth AL. Open randomised trial of prescribing strategies in managing sore throat. Br Med J 1997; 314: 722-7.
  6. MacFarlane JT, Holmes WF, MacFarlane RM. Reducing reconsultations for acute lower respiratory tract illness with an information leaflet: a randomised controlled study of patients in primary care. Br J Gen Pract 1997: 47: 719-22.
  7. Evans RS, Pestotnik SL, Classen DC. A computer-assisted management program for antibiotics and other anti-infective agents. N Engl J Med 1998; 338: 232-8.
  8. Kasanen A, Hajba A, Junnila SYT, Sundquist H. Comparative study of trimethoprim and cephalexin in urinary tract infection. Curr Ther Res, Clin Exp 1981; 29: 477-85.
  9. Baraff LJ, Ablon WD. Cefaclor versus ampicillin for outpatient treatment of urinary tract infections. Am J Emerg Med 1984; 2: 327-30.
  10. Fahey T, Stocks N, Thomas T. Quantitative systematic review of randomised controlled trials comparing antibiotic with placebo for acute cough in adults. Br Med J 1998; 316: 906-10.
  11. Magee JT, Pritchard EL. Antibiotic prescribing and antibiotic resistance in community practice: retrospective study, 1996-8. Br Med J 1999; 319: 1239-40.

NHS Beacon Awards
Beacon status is awarded to practices, trusts and other healthcare organisations within the NHS that have demonstrated good practice. The NHS Beacon programme aims to spread best practice across the health service. For further information visit the website: www.nhs.uk/beacons


Guidelines in Practice, June 2001, Volume 4(6)
© 2001 MGP Ltd
further information | subscribe