Has UK guidance affected general practitioner antibiotic prescribing for otitis media in children?
Journal of Public Health |
Has UK guidance affected general practitioner antibiotic prescribing for otitis media in children?
P. L. Thompson 2
R. E. Gilbert 1
P. F. Long 0
S. Saxena 4
M. Sharland 3
I. C. K. Wong 2
0 Department of Pharmaceutics and Centre for Paediatric Pharmacy Research, The School of Pharmacy , 29-39 Brunswick Square, London WC1N 1AX , UK
1 Centre for Evidence-based Child Health, Institute of Child Health , London WC1N 1EH , UK
2 Centre for Paediatric Pharmacy Research, The School of Pharmacy, University of London , BMA House, Tavistock Square, London WC1H 9JP , UK
3 Paediatric Infectious Diseases Unit, St. George's Hospital , London SW17 0RE , UK
4 Department of Primary Care and Social Medicine, Imperial College London , London W6 8RP , UK
A B S T R AC T Background Since 1997, UK guidance has advocated limiting antibiotic prescribing for otitis media. It is not known whether this has influenced general practitioner prescribing practice. Aims and objectives To investigate the trends in diagnoses and antibiotic prescribing for otitis media in children in relation to guidance. Methods We used the General Practice Research Database to conduct time-trend analyses of diagnoses and antibiotic prescribing for otitis media in 3 months to 15 years old, between 1990 and 2006. Results A total of 1 210 237 otitis media episodes were identified in 464 845 children; two-thirds (68%; 818 006) received antibiotics. Twentytwo percent (267 335) were classified as acute, 85% (227 335) of which received antibiotics. Overall, antibiotic prescribing for otitis media declined by 51% between 1995 and 2000. Much of this reduction predated guidance. During this period, prescribing for otitis media coded as acute increased by 22%. Children diagnosed with acute otitis media were more likely to receive antibiotics than otitis media not coded as acute (P , 0.05). From 2000 prescribing plateaued, despite publication of further guidance. Otitis media diagnoses consistently paralleled prescribing. Conclusions The reduction in antibiotic prescribing for otitis media predated guidance. The simultaneous decrease in prescribing for non-acute otitis media and increase for acute otitis media suggest diagnostic transfer, possibly to justify the decision to treat.
children; ear disorders; epidemiology
Otitis media is a common childhood infection diagnosed in
12% of all children consulting their general practitioner
(GP)1 and accounting for 14% of all antibiotic prescriptions
Since 1997, a series of systematic reviews, government
reports and evidence-based clinical practice guidelines have
been published in the UK covering the management of otitis
media.3 – 8 All reached similar conclusions that there is a
limited role for antibiotics in the treatment of otitis media.
Acute otitis media is a self-limiting condition with around
80% of episodes spontaneously resolving within 2 to 7 days
without antibiotic treatment3 and antibiotics are rarely justified
for non-acute otitis media. Since 2003, guidance has
specifically advocated delayed prescribing (where an antibiotic
prescription is written only for use if the patient is not improving
after 48 – 72 h), particularly in children over 2 years of age.6 – 8
However, it is not known whether this guidance has had any
impact upon GP antibiotic prescribing practice.
The aim of this study was to investigate the trends in
diagnoses and antibiotic prescribing for otitis media in children,
in relation to the UK guidance covering the treatment of
otitis media. iCAP (the improving Children’s Antimicrobial
Prescribing group) used routinely recorded GP consultation
and antibiotic prescription data to examine these trends.
We conducted time-trend analyses of diagnoses and antibiotic
prescribing for otitis media using routinely recorded data
from the General Practice Research Database (GPRD).9,10
Analyses were based on all children aged 3 months to
15 years registered with an ‘up-to-standard’ GPRD practice11
between 1 January 1990 and 31 December 2006.
The GPRD is one of the world’s largest computerized
databases of anonymised longitudinal general practice patient
records, comprising 6% of the UK census population.
Practices contributing to the GPRD are representative of
practices in the total UK and are under contract to record all
diagnoses, prescriptions, immunisations, hospital referrals and
test results of all active patients.9,10 Validation studies show
quality and completeness of the GPRD data are high.12
Identification of otitis media episodes and
antibiotics prescribed for otitis media
Diagnoses in the GPRD are classified using either Oxford
Medical Information Systems (OXMIS) or Read codes. We
identified children with otitis media by searching the GPRD
for diagnostic codes corresponding to otitis media or a
related label of middle ear disease (Appendix A, available
online as supplementary data). Diagnoses were reassigned to
sub-groups, classified as either acute or non-acute otitis
media; sign and symptom codes indicating acute infection,
such as ear discharge or earache, were classified as acute
otitis media. Otitis media refers to the acute plus non-acute
otitis media diagnoses combined. As otitis media can recur,
a 14-day screening period was set from the initial otitis
media record during which time-related events were
classified as the same episode; the first diagnosis after this period
was once again counted as an incident episode and the cycle
continued. Any antibiotic listed under the British National
Formulary Section 5.1 (with the exception of
antituberculosis and anti-leprotic drugs) prescribed in the same
consultation as otitis media was diagnosed were classified as
being prescribed for otitis media.
We calculated age- and calendar year-specific incidence rates
for otitis media diagnoses and for antibiotic prescribing for
otitis media, per 1000 child years at risk in the GPRD. Data
are given for children aged less than and greater than 2 years
as some UK guidance differs for these age groups (Box 1).7,8
95% confidence intervals (95% CIs) were generated using
Box 1 Timeline showing the publication of
systematic reviews, government reports, clinical
practice guidelines and other UK national
events influencing GP antibiotic prescribing
practice for otitis media
(1) October 1992: Haemophilus influenzae type b vaccine
(2) 1995: Vision software introduced for GPRD data
(3) 1995: READ codes introduced for diagnostic classification
in the GPRD15
(4) January 1997: Cochrane Library Review (Glasziou
et al.)—concluded that antibiotics are of limited benefit
for most children with acute otitis media, recommended
(5) December 1997: NHS Direct Helpline introduced16
(6) April 1998: House of Lords Select Committee on Science
and Technology report into antibiotic resistance called for
guidelines on limiting antibiotic use4
(7) September 1998: All GPs sent the Standing Medical
Advisory Committee report stating that many cases of
acute otitis media do not require antibiotics and
advocating a national campaign on antibiotic treatment5
(8) September 1999: National Advice to the Public campaign
to educate the public of the need to reduce antibiotic
(9) February 2003: SIGN (Scottish) e-guidelines—
recommended analgesics for acute otitis media, with
5-day amoxicillin if no improvement after 24 – 72 h6
(10) June 2004: PRODIGY (NHS, UK) e-guidelines—
recommended analgesics for acute otitis media, with the
potential for 5-day amoxicillin in children under 2 years
(11) September 2006: Prevenar vaccine introduced18
(12) October 2006: Meta-analysis in The Lancet (Rovers
et al.)—recommended delayed antibiotic prescribing for
acute otitis media in children over 2 years of age8
Poisson approximation and tests for linear trend were
conducted. Data management and analyses were performed
using Stata software version 9.2.13
Search for UK clinical practice guidance covering
the treatment of otitis media
We identified UK systematic reviews, government reports
and clinical practice guidelines covering the treatment of
otitis media by searching the Cochrane Central Register of
Controlled Trials (CENTRAL) (www.cochrane.org),
Evidence-Based Medicine for Primary Care and Internal
Medicine (www.ebm.bmj.com), National Library for Health
(www.library.nhs.uk), National Library for Medicine (www.
nlm.nih.gov), National Institute for Health and Clinical
Excellence (NICE) (www.nice.org.uk), Turning Research
Into Practice (TRIP) database (www.tripdatabase.com) and
hand-searching reference lists of appropriate articles.
Relevant data on the chronology and key recommendations
of these documents were extracted. Individual trials were
not included as it was considered unlikely that these would
have had a national influence on GP prescribing practice.
Between 1990 and 2006 there were a total of 2 622 348
children aged 3 months to 15 years within the GPRD,
contributing 7 119 677 child years of follow-up data from 423
general practices. A total of 1 210 237 episodes of otitis
media were identified in 464 845 children. Approximately,
half (51%; 617 221) of the episodes occurred in boys. The
mean age at otitis media diagnosis was 5.4 (SD + 3.8) years.
Incidence of otitis media peaked in children aged ,1 year
and just over one-fifth (21%; 258 638) of all otitis media
episodes occurred in children under 2 years of age. Children
had an average of 0.46 episodes of otitis media per year,
decreasing with increasing age (P , 0.05).
Two-thirds of all otitis media episodes (68%; 818 006)
received an antibiotic prescription. Twenty-two percent
(267 335) of all otitis media episodes were classified as
acute otitis media, 85% (227 335) of which received an
antibiotic. A child diagnosed with acute otitis media was
significantly more likely to receive an antibiotic
prescription than a child with otitis media not coded as acute
(85% versus 63% treated, respectively; P , 0.05).
Amoxicillin accounted for 76% (622 339) of all antibiotics
prescribed for otitis media, whilst erythromycin was the
second most commonly prescribed antibiotic (15%; 122
701). There were no differences in the choice of antibiotic
between children diagnosed with acute otitis media and
those with non-acute otitis media.
Fig. 1 shows that antibiotic prescribing for otitis media in
children under 2 years of age increased by 32% between
1990 and 1993 (from 350.9 (95% CI: 344.0 – 357.7) to
463.3 (95% CI: 457.3 – 469.2) prescriptions per 1000 child
years; P ¼ 0.09), while rates remained stable in the older age
groups over this time period. From 1995 until 2000,
antibiotic prescribing for otitis media decreased significantly in
all age groups (overall decline ¼ 51%, from 177.4 (95% CI:
176.1 – 178.7) to 86.7 (95% CI: 86.0 – 87.5) prescriptions per
1000 child years; P , 0.05) and plateaued thereafter. The
proportion of otitis media episodes treated with antibiotics
decreased significantly over the study period, for all otitis
media diagnoses and all age groups (overall treated ¼ 77%
in 1990 versus 58% in 2006; P , 0.05; Fig. 2).
Fig. 2 Annual incidence of diagnoses and antibiotic prescribing for all otitis media in children (aged 3 months to 15 years) in UK general practice (depicting
the percentage of otitis media episodes treated with antibiotics and events from Box 1).
SR ¼ systematic review, GR ¼ government report, HPC ¼ health promotion campaign, G ¼ guideline.
Six UK documents covering the treatment of otitis media
were identified and are summarized in Box 13 – 8, along with
other significant events likely to influence GP antibiotic
prescribing nationally. Fig. 1 shows that the reduction in
antibiotic prescribing for otitis media predated the first UK
systematic review that reported the limited benefit of
antibiotics for otitis media (Cochrane Collaboration review by
Glasziou et al.3); over one-third (39%) of the total decrease
in antibiotic prescribing observed between 1995 and 2000
had already occurred prior to 1997. Antibiotic prescribing
for otitis media plateaued from 2000 onwards in all age
groups despite the publication of further national guidance
advocating limiting antibiotic prescribing for otitis media.
Fig. 1 shows contrasting trends in antibiotic prescribing
for acute and non-acute otitis media. Between 1995 and
2000, a 64% decrease in antibiotic prescribing for non-acute
otitis media (from 149.8 (95% CI: 148.6 – 151.1) to 53.2
(95% CI: 52.6 – 53.8) prescriptions per 1000 child years;
P , 0.05) coincided with a 22% increase in antibiotic
prescribing for acute otitis media (from 27.5 (95% CI:
27.0 – 28.1) to 33.6 (95% CI: 33.1 – 34.1) prescriptions per
1000 child years; P , 0.05). This diagnostic transfer was
most striking in children aged ,2 years; 61% decrease in
antibiotic prescribing for non-acute otitis media (from 391.9
(95% CI: 386.2 – 397.5) to 151.5 (95% CI: 148.1 – 155.0)
prescriptions per 1000 child years; P , 0.05) accompanied
by a 45% increase in antibiotic prescribing for acute otitis
media (from 74.2 (95% CI: 71.2 – 77.2) to 107.6 (95% CI:
104.6 – 110.5) prescriptions per 1000 child years; P , 0.05).
Otitis media diagnoses paralleled antibiotic prescribing for
otitis media for all age groups throughout the study period,
for both acute and non-acute diagnoses (Fig. 2. See also
Table 1, available online as supplementary data).
Main findings of this study
Antibiotic prescribing for otitis media halved between 1995
and 2000. Much of this fall predated the emergence of UK
guidance advocating limiting antibiotic treatment for this
condition. The overall decrease in antibiotic prescribing for
otitis media coincided with an increase in prescribing for
otitis media coded as acute, most marked in children
aged ,2 years. GPs were more likely to code otitis media
episodes as acute if antibiotics were prescribed.
What is already known on this topic
Our findings are consistent with a previous study reporting
falls in antibiotic prescribing for otitis media between 1991
and 2000, for all ages, with opposing trends for otitis media
labelled as acute and that as glue ear (otitis media with
effusion).19 However, it is not known whether the UK guidance
advocating limiting antibiotic use for otitis media has had
any impact upon GP prescribing practice.
What this study adds
Our study explains the trends in diagnoses and antibiotic
prescribing for otitis media in relation to the UK guidance
covering the treatment of this condition. Additionally, the
inclusion of more recent data, until 2006, shows that
incidence rates of diagnoses and antibiotic prescribing for otitis
media have stabilized since the year 2000.20
Antibiotic prescribing for otitis media was already in steep
decline prior to the 1997 publication of the first UK
systematic review reporting the limited benefit of antibiotics for otitis
media3 and plateaued from 2000 onwards despite the
publication of further UK national guidance. Evidence of poor
compliance with post-2000 UK guidelines advocating limiting
antibiotic use for otitis media21 raises the question of whether
GPs have reached the lowest level of antibiotic prescribing
that they find acceptable. In an attempt to assess how much
lower antibiotic prescribing could go, we compared the rates
of antibiotic prescribing for otitis media in the UK with the
Netherlands,22 the country with the lowest rate of antibiotic
prescribing in the world23,24 (Fig. 3). Despite the 51%
reduction in UK antibiotic prescribing for otitis media
between 1995 and 2000, prescribing rates in the UK are still
more than twice that of the Netherlands (2003 ¼ 164.0
versus 72.5 prescriptions per 1000 child years, respectively).
This suggests that UK antibiotic prescribing for otitis media
could most likely be safely reduced even further.
Limitations of this study
This is the largest study to date investigating the trends in
diagnoses and antibiotic prescribing for otitis media in
children. Strengths of our study include the large sample size, the
high quality and completeness of GPRD data12 and the
representiveness of general practices in the GPRD to those of the
total UK population, where virtually all antibiotic prescribing
is initiated.5 The national representiveness of our study is
confirmed by the consistency of findings from other UK
wide studies with our results.25,26 A further strength was our
use of a comprehensive approach to capture all Read/OXMIS
codes indicative of otitis media, in an attempt to minimize the
problems associated with coding variability between GPs. By
then sub-classifying the diagnoses into acute and non-acute
otitis media, we revealed opposing trends in diagnoses and
antibiotic prescribing in the younger age groups.
However, our study has a number of important limitations.
The major weakness, common to all studies based on
routinely collected clinical data, is misclassification of the reason
for consultation or treatment. Our assumption that an
antibiotic prescribed in the same GP consultation as otitis media
was diagnosed indicated that the antibiotic was prescribed for
otitis media may have been wrong. If an antibiotic was given
for another ailment, this would have led to an overestimation
of antibiotic prescribing for otitis media. Similarly, using
GPRD data, it was not possible to test for associations
between the publication of guidance and GP prescribing
practice as the motives for prescribing are unknown. However,
the suggested reasons for the changes in antibiotic prescribing
for otitis media are supported by the existing literature in this
Patient adherence data are not available from the GPRD.
Thus, it was not possible to investigate whether GPs were
practising delayed prescribing, whereby an antibiotic
prescription may have been issued but the GP requested for the
patient/parent not to redeem it unless symptoms persisted
after 48 – 72 h, as recommended by the SIGN guidelines6,
PRODIGY guidelines7 and the meta-analysis by Rovers et al.8
This could have led to an over-estimation of the proportion of
otitis media episodes that were treated with antibiotics.
Additionally, GPs contributing data to the GPRD are asked
to record ‘all significant diagnoses’12 and may
underreport untreated otitis media, thereby, also leading to an
overestimation of antibiotic treated otitis media. A further problem
was the data did not allow assessment of severity of otitis
media; it may be that antibiotic prescribing for sicker children
has remained stable while inappropriate prescribing has
diminished.29 GPs may also have different thresholds for
prescribing, which could impact upon the results obtained, but these
data serve to reflect prescribing as it stands in general practice.
Finally, the findings presented in this study were derived from
the population consulting their GP. Ideally, to examine otitis
media trends, the incidence of otitis media in the entire
population, whether consulting their GP or not, is required.
However, it is effectively impossible to collect this and GP
consultation data, therefore, provides the most accurate
measure of otitis media incidence as is practically achievable.30
The reasons for the trends are not known. An increase in
overall antibiotic prescribing from 1980 until 199531,32
followed by a reduction between 1995 and 2000 has previously
been reported.2,33,34 The decline in antibiotic prescribing for
otitis media is likely to reflect a true reduction in the overall
occurrence of otitis media as it is associated with a decrease
in GP consultations for respiratory tract infections.25,26 The
decline is unlikely to be explained by changes in
healthcareseeking behaviour as overall GP consultation rates in
children remained stable between 1995 and 2000.35,36
The simultaneous decrease in diagnoses and antibiotic
prescribing for non-acute otitis media and increase for acute
otitis media suggest diagnostic transfer, possibly to justify
the decision to treat. This is further supported by the
consistent parallel trends for diagnoses and prescribing and
the fact that GPs were more likely to code the otitis media
episode as acute if an antibiotic was prescribed. Such
interpretation is plausible as disease is ultimately classified
at the discretion of each GP and marked variation in the
recording of otitis media between GPRD practices has
previously been reported.37 As early as 1972, Howie27
proposed that the diagnosis may at times be a justification for
treatment, rather than the reason for it and Mangione-Smith
et al.28 reported that the perceived parental pressure to
prescribe an antibiotic was associated with the diagnosis given,
possibly as a means of rationalizing the treatment decision.
The effects of accountability for prescribing, which may
increase the need to come up with a diagnosis that justifies
the treatment decision, have not yet been explored.
The reasons why GPs may not adhere to the guidance are
not fully understood and cannot be determined from this
study. However, it is known from previous studies27,38 – 42 that
the decision to treat goes beyond pharmacological reasoning
and is based on the interaction between the prescriber
and patient with both medical and non-medical factors
influencing the prescribing decision, such as fears over the
development of complications29 or parental pressure for an
antibiotic.28,29,43 – 45 Non-medical motives are especially true
for young children and this may explain the marked increase
in antibiotic prescribing for acute otitis media in children
aged ,2 years.
Qualitative research is needed to examine what motivated
the changes in prescribing and to understand why, despite
the evidence-based guidance, GP prescribing for otitis
media has changed very little since 2000. Formal monitoring
of national rates of disease incidence, antibiotic prescribing
and potential complications is required to assess the impact
and ensure the safety of further national healthcare
guidance. In July 2008, the National Institute for Health and
Clinical Excellence issued a clinical guideline entitled
‘Prescribing of antibiotics for self-limiting respiratory tract
infections in adults and children in primary care’,
recommending a delayed or no prescribing strategy for all upper
respiratory tract infections, including otitis media.46 Will this
guideline further reduce the use of antibiotics for otitis
media? Our findings suggest that further guidance urging
GPs to reduce antibiotic prescribing is unlikely to have a
major impact unless combined with effective
implementation strategies. It is essential that prescribers are made
aware of their reasons for prescribing, particularly their
nonmedical motives, and that the barriers to implementation are
more clearly identified, understood and addressed when
putting future guidance into place, otherwise it seems
destined to fail.
I.C.K.W.’s post was funded by the Department of Health
Public Health Career Scientist Award. S.S. holds a National
post doctoral award from the Department of Health. The
license for the GPRD was funded by the European
Commission via the Taskforce in Europe for Drug
Development for the Young (TEDDY) network of Excellence
European Commission Framework 6 Programme, 2005 – 10.
All researchers were independent from the funders.
We thank the general practitioners who contributed data to
R.G. had the original idea for the study. P.L.T. extracted the
relevant data from the GPRD. P.L.T. and RG developed
the analytical strategy and all authors were involved in the
interpretation of the data. P.L.T. prepared the first draft of
the manuscript and all authors contributed to writing
subsequent and final drafts. I.C.K.W. is the paper guarantor.
P.L.T., R.G., M.S. and I.C.K.W. were members of the
Department of Health’s Specialist Advisory Committee on
Antimicrobial Resistance (SACAR), paediatrics sub-group.
P.F.L. and S.S. have no competing interests.
Ethics approval for this research
the Scientific and Ethical Advisory
was obtained from
Supplementary data are available at Journal of Public Health
Glasziou PP, Del Mar CB, Sanders SL et al. Antibiotics for acute
otitis media in children (Cochrane Review). Cochrane Database of
Systematic Reviews, 1997, Issue 1. Art. No.: CD000219. DOI:
House of Lords Select Committee on Science and Technology.
Resistance to antibiotics and other antimicrobial agents, 7th report
1997 – 1998, The Stationery Office, London, 1998.
1 Freid VM , Mukuc DM , Rooks RN . Ambulatory health care visits by children: principal diagnosis and place of visit . Vital Health Stat 1998 ; 13 : 1 - 23 .
2 Thompson PL , Spyridis N , Sharland M et al. Changes in clinical indications for community antibiotic prescribing for children in the UK from 1996 - 2006: will the new NICE prescribing guidance on upper respiratory tract infections be ignored? Unpublished data .
5 Standing Medical Advisory Committee . The Path of Least Resistance. London: Department of Health , 1998 . www.advisorybodies. doh. gov.uk/SMAC/SMAC1.HTM (accessed 14 July 2008 ).
6 SIGN (Scottish Intercollegiate Guidelines Network). Diagnosis and management of childhood otitis media in primary care . A national clinical guideline , 2003 . www.sign. ac.uk/guidelines/fulltext/66/ index.html (accessed 14 July 2008 ).
7 PRODIGY. Otitis media-acute , 2004 . www.prodigy. nhs.uk/otitis_ media_acute (accessed 14 July 2008 ).
8 Rovers MM , Glasziou P , Appelman CL et al. Antibiotics for acute otitis media: a meta-analysis with individual patient data . Lancet 2006 ; 368 : 1429 - 35 .
Lancet 1997 ; 350 : 1097 - 9 .
10 Wong IC , Murray ML . The potential of UK clinical databases in enhancing paediatric medication research . Br J Clin Pharmacology 2005 ; 59 : 750 - 5 .
11 Card T , Logan RFA , Rodrigues LC et al. Antibiotic use and the development of Crohn's disease . Gut 2004 ; 53 : 246 - 50 .
12 Hollowell J. The general practice research database: quality of morbidity data . Population Trends 1997 ; 87 : 36 - 40 .
13 www.stata. com (accessed 14 July 2008 ).
14 Joint Committee on Vaccination and Immunisation. JCVI Statement: Haemophilus influenzae type b (Hib) Disease and Hib Vaccine. www.advisorybodies.doh.gov.uk/JCVI/hib.pdf (accessed 14 July 2008 ).
15 Medicines and Healthcare Products Regulatory Agency. GPRD-an overview , 2004 . Crown Copyright.
16 Department of Health. The New NHS: Modern and Dependable , 1997 . www.archive. official-documents.co.uk/document/doh/newnhs/ contents.htm (accessed 14 July 2008 ).
17 Department of Health. Antibiotics: don't wear me out , 1999 . www. dh. gov.uk/PublicationsAndStatistics/Publications/Publications PolicyAndGuidance/PublicationsPolicyAndGuidanceArticle/fs/en? CONTENT_ID¼4007460&chk¼VHEgjn (accessed 14 July 2008 ).
18 Department of Health Letter. Planned changes to the routine childhood immunisation programme , 2006 . www.info.doh.gov.uk/doh/ embroadcast.nsf/0/D474DD433E03C7F38025710F003B450B/$File/ Changes%20to% 20childhood%20immunisation%20Programme% 208%20Feb%2006 .rtf ? OpenElement (accessed 14 July 2008 ).
19 Williamson I , Benge S , Mullee M et al. Consultations for middle ear disease, antibiotic prescribing and risk factors for re-attendance: a case-linked cohort study . Br J Gen Pract 2006 ; 56 : 170 - 5 .
20 Thompson PL , Murray ML , Sharland M et al. Up-to-date findings show change in acute otitis media consultation trend [Letter] . Br J Gen Pract 2006 ; 526 : 379 - 80 .
21 Woolley SL , Smith DRK. Acute otitis media in children-there are guidelines but are they followed ? J Laryngology Otology 2005 ; 119 : 524 - 8 .
22 Plasschaert AIO , Rovers MM , Schilder AGM et al. Trends in doctor consultations, antibiotic prescription, and specialist referrals for otitis media in children: 1995 - 2003 . Pediatrics 2006 ; 117 : 1879 - 86 .
23 Froom J , Culpepper L , Grob P et al. Diagnosis and antibiotic treatment of acute otitis media: report from International Primary Care Network . Br Med J 1990 ; 300 : 582 - 6 .
24 Goossens H , Ferech M , Stichele RV et al. for the ESAC Project Group. Outpatient antibiotic use in Europe and association with resistance: a cross-national database study . Lancet 2005 ; 365 : 579 - 87 .
25 Fleming DW , Ross AM , Cross KW et al. The reducing incidence of respiratory tract infection and its relation to antibiotic prescribing . Br J Gen Pract 2003 ; 53 : 778 - 83 .
26 Ashworth M , Latinovic R , Charlton J et al. Why has antibiotic prescribing for respiratory illness declined in primary care? A longitudinal study using the general practice research database . J Public Health 2004 ; 26 : 298 - 74 .
27 Howie JGR . Diagnosis: the archilles heel . J R Coll Gen Pract 1972 ; 22 : 310 .
28 Mangione-Smith R , McGlynn EA , Elliott MN et al. The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behaviour . Pediatrics 1999 ; 103 : 711 - 18 .
29 Kumar S , Little P , Britten N. Why do general practitioners prescribe antibiotics for sore throat? Grounded theory interview study . Br Med J 2003 ; 326 : 138 .
30 Fleming DM , Ross AM , Cross KW et al. The reducing incidence of respiratory tract infection and its relation to antibiotic prescribing . Br J Gen Pract 2003 ; 53 : 778 - 83 .
31 Davey PG , Warren RE , Bax RP et al. Growth in the use of antibiotics in the community in England and Scotland in 1980 - 93 . Br Med J 1996 ; 312 : 613 .
32 Prescription Pricing Division . Prescribing review-antibacterial drugs , 2007 . http://www.ppa. org.uk/news/pact-122006.htm (accessed 16 July 2008 ).
33 Wrigley T , Tinto A , Majeed A. Age-, sex-specific antibiotic prescribing patterns in general practice in England Wales , 1994 to 1998 . Health Stat Q 2002 ; 14 : 14 - 20 .
34 Sharland M , Kendall H , Yeates D et al. Changes in antibiotic prescribing in general practice and hospital admissions for quinsy, mastoiditis and rheumatic fever , in children. Br Med J 2005 ; 331 : 328 - 9 .
35 Office for National Statistics. Living in Britain: General Household Survey 2002 , Chapter 7. General health and use of health services . London: ONS, 2002 , pp. 109 - 113 . www. statistics.gov.uk/lib2002/downloads/health.pdf (accessed 14 July 2008 ).
36 Office of Health Economics . Compendium of Health Statistics, 17th edn , 2005 - 2006 . (London: OHE), 2006 .
37 De Wilde S , Carey IM , Bremner SA et al. A comparison of the recording of 30 common childhood conditions in the Doctors' Independent Network and General Practice Research Databases . Health Stat Q 2004 ; 22 : 21 - 30 .
38 Stephenson MJ , Henry N , Norman GR . Factors influencing antibiotic use in acute respiratory tract infections in family practice . Can Family Phys 1988 ; 34 : 2149 - 52 .
39 Bradley CP . Uncomfortable prescribing decisions: a critical incident study . Br Med J 1992 ; 304 : 294 - 6 .
40 Armstrong D , Reyburn H , Jones R. A study of general practitioner's reasons for changing their prescribing behaviour . Br Med J 1996 ; 312 : 949 - 52 .
41 Butler C , Rollnick S , Pill R et al. Understanding the culture of prescribing: qualitative study of general practitioners' and patients' prescriptions of antibiotics for sore throat . Br Med J 1998 ; 317 : 637 - 42 .
42 Damoiseaux RAMJ , de Melker RA , Ausems JE et al. Reasons for non-guideline-based antibiotic prescriptions for acute otitis media in The Netherlands . Family Pract 1999 ; 16 : 50 - 3 .
43 Coenen S , Michiels B , Renard D et al. Antibiotic prescribing for acute cough: the effect of perceived patient demand . Br J Gen Pract 2006 ; 56 : 183 - 90 .
44 Britten N , Ukoumunne O. The influence of patient's hopes of receiving a prescription on doctor's perceptions and the decision to prescribe: a questionnaire survey . Br Med J 1997 ; 315 : 1506 - 10 .
45 MacFarlane J , Holmes W , MacFarlane R et al. Influence of patients' expectations on antibiotic management of acute lower respiratory illness in general practice: questionnaire study . Br Med J 1997 ; 315 : 1211 - 4 .
46 National Institute for Health and Clinical Excellence . Prescribing of antibiotics for self-limiting respiratory tract infections in adults and children in primary care . www.nice.org.uk/guidance/index.jsp? action¼byID&o¼11816 (accessed 14 July 2008 ).